Light Treatment Of Hypothyroidism – Is It Possible With LEDs?

Light Therapy For Hypothyroidism


The purpose of this article is to look at the published data in this field (photobiomodulation in the treatment of hypothyroidism) and see how this could be applied to relatively cheap LEDs. Many people have taken to using these devices but it seems that most don’t take a lot of notice of the data from the studies. Dosing is a key factor in photobiomodulation and more is not necessarily better.

Photobiomodulation is one of the terms used for treatments using low levels of red or near infrared light. This article is designed to look at some of the metrics and see if it is possible to use home devices to treat hypohtyroidism, and if so how to closely design the dosing parameters to match with the successful studies. Many leaps must be made because different hardware is used in the studies, thin laser beams or pulsed high power lasers. There is no doubt that LEDs work in general for photobiomodulation, there is doubt about how to translate dosing data from laser to LED devices. I’m not a doctor and these are not recommendations for treating your thyroid. My only recommendation is to read the data yourself before applying light to your thyroid or other sensitive tissues.


Evidence For Photobiomodulation In The Treatment Of Hypothyroidism

A number of human studies published between 1996 and 2017 have shown that light applied to the thyroid or other parts of the body can significantly improve the following aspects of hypothyroidism and associated autoimmunity in humans:

  • Free T3, total T3, free T4, total T4
  • TSH, anti-TPO (TPOab)
  • Goiter
  • Markers of cellular immunity (T cells)
  • Ultrasound results, Cyst size
  • Sensation of thyroid compression in the throat
  • Reduction of facial swelling and other oedema
  • Soften thyroid tissue (on palpation)
  • Lower TGF-b1
  • Reversal of dyslipidaemia (the changes in blood lipid markers often associated with the development of heart disease)

Over the course of these experimental treatments many people have been able to lower their medication and quite a few have gone into remission. The procedures, when done correctly, seem to nudge the thyroid into a corrective state where it often stays for 3-6 months before requiring further application.

You can access my categorized spreadsheets of thyroid related photobiomodulation studies on my Patreon page by clicking this link. Please consider becoming a patreon supporter if you find these articles useful.

Can Light Therapy Do Damage When Dosed Incorrectly?

There are reasons to be cautious. While I do not know of any negative thyroid studies in humans, there are some in animal models. One rabbit study showed a decrease in thyroid hormones T3 and T4 and a concurrent increase TSH – test results indicating a movement towards a hypothyroid state.


There’s also animal data relating to correction of hyperthyroidism with light. This animal data showing a lowering of hormones in normal animals suggests that light can damage thyroid function in healthy animals (rather than assuming light merely corrects hyperthyroid or hypothyroid states).

The negative dose parameter used in the rabbit study has not been used in humans (to my knowledge). Positive human studies are in the same range as positive rabbit studies, so it might be best to avoid this particular metric. Specifically 150mW lowered thyroid hormones in rabbits, whereas 50mW increases hormones in humans and animals. I’ll speak more about dosing further on in the article. For more on this, see my article:

Light And Hypothyroidism – Cure Or Cause? 

Harm From Light Therapy

There are also number of other animal studies showing negative effects in sensitive tissues. A couple of studies show negative effects in testicle function and tissues from certain doses of light.

“In the 670 nm wavelength group, serum T level was also significantly increased the testosterone levels at the same intensity of 360 J/cm /day. On histopathological examination, there were no definite changes in the 670 nm wavelength group. In the 808 nm wavelength group, there were such findings as an atrophy of the seminiferous tubules, disarrangement of sertoli cells, generation of giant multinucleated bodies and other deformities. It is apparent that the adverse effects occurred but serum T level was not increased following a 5- day course of irradiation with an 808-nm wavelength at intensity of 360 J/cm /day. Taha MF et al. showed similar histopathological findings. According to these authors, on histopathological examination of the testis irradiated with an 830-nm wavelength at lower doses (28.05J/cm ), there were normal appearances of the seminiferous epithelium and interstitial tissue. These authors also noted, however, that the seminiferous epithelium and interstitial tissue were irregularly arranged following the irradiation with an 830-nm wavelength at higher doses (46.8 J/cm ).”(*)
************************“Stimulation of aerobic phosphorylation by LLLT may have led to a deregulated increase in ROS leading to sperm damages. Thus, LLLT at energy of 28 J/cm2 (808 nm of wavelength and 30 mW of power output) can induce sperm damages and increase the quantity of cells in seminiferous tubule in rams.”(*)

Point being that inappropriate dosing could do damage.


Discovery Of Low Level Laser Therapy

While attempting to cut tumors from experimental animals Endre Mester was provided with an early model of laser which had been mis-calibrated to produce a much lower energy output than requested. The resulting 5mW laser beam had no capacity to cut, but after the treatment Mester observed accelerated wound healing and hair growth in the experimental animals. For more on the development of light therapy see the history of red light (*)Since then it has been shown that “low” is a critical component of photobiomodulation, and it’s very easy to use too high of a dose.

Biphasic Dose Response And The Problem Of Overdosing

Low level laser therapy, cold laser therapy, low level light therapy and photobiomodulation are the most common names for medical treatments of this sort. Since that initial discovery in the 1960s many observations have made about the range of parameters in which this type treatment works effectively. As you can see from these terms it is implied that the therapy be low level (meaning low energy) and “cold” (low energy in a certain wavelength range). There is a well-known biphasic dose response in photobiomodulation (PBM).

Biphasic Dose Response in Low Level Light Therapy – An Update


The biphasic means that there is a therapeutic dose range surrounded at the upper and lower extremes by an ineffective dose range. You can easily “overdose”. Using dose parameters that are excessive usually have no beneficial effect. However, I think some of the studies above show that there is reason to be concerned about damage from excessive dosing in sensitive tissues. Heat is generally seen as something to be avoided in PBM, but a couple of the studies showing damage from higher doses speculate that these negative effects might be from increasing reactive oxygen species (ROS). This mechanism seems viable to me. I assume that heat is not the only potential problem with inappropriate dosing.

Possible Explanations For Biphasic Dose Response In LLLT

Low Level Light Therapy Dosing Parameters In Thyroid Studies




Wavelength is measured in nanometres (nm). For visible light – wavelength will determine its colour. The field of photobiomodulation generally uses light sources that have a narrow band of wavelengths in the region of 600nm to 1100 nm. Those light sources are laser and LED. Other light sources are too wide in spectrum and often include heat generating wavelengths, with heat you cannot safely get enough power in the 600-1100nm range. Heat bulbs and incandescents are useful lights, but not suitable for these purposes. The lower end of this 600-1100nm spectrum is visible red light. Somewhere around 780 nm light becomes invisible to the naked eye and enters the range that we call the infrared.

As you go up in wavelength through the infrared spectrum you start to reach wavelengths that  generate more heat, the wavelengths therapeutic “heat lamps” are based on. It is only the region of infrared closest to red that is used in  for photobiomodulation. This region generates relatively little heat compared to further up the IR spectrum but has greater penetration than visible red. This portion of the infrared spectrum, from about 780 nm to about 1100 nm is called near-infrared or near-infrared A (NIRA-IRA).


Biological Peaks Within The Red And NIRA Range

Some researchers have tested the biological effects of different wavelengths within this spectrum (600-1100nm). Some peaks with higher biological activity have been identified. Some of these important peaks are around 620 mm, 680 nm, 760 mm, and 820 nm. Wavelengths close to these seem to be more active on the mitochondria, affecting cytochrome C oxidase and a host of other factors. This is not to say that other wavelengths between 600nm and 1100nm are not useful, but that in general we might consider them less potent or desirable, except in instances where proven otherwise.



Power Is A Critical Factor

The power output of a light source is measured in watts(W) or milliwatts(mW). There are 1000 mWs in a Watt. Endre Mester’s original discoveries on the rejuvenating effect of red light were with a laser that had an output of approximately 5 mW. Most of the beneficial studies that I’ve read tend to have a power output between 10mW and 150 mW. Effective studies in humans seem to cluster more aggressively when you limit the range to 20mW to 100 mW. There are some positive studies using higher powered light (there are exceptions to almost every generalisation I will make in this article).


Most of the positive studies with higher power light sources are pulsed. Pulsing involves turning the light source on and off multiple times per second. One of the reasons for pulsing is to allow for the dissipation of heat at a higher peak power output. Higher peak power output allows for deeper penetration and access to tissues that would otherwise not be accessible by light. Devices that have pulsing patterns in the Watt range range are quite expensive at this point and so I’m not going to consider them in this article. See my previous articles and database for more on those. You should not consider pulsed studies as the same as non pulsed studies. 5,000mW pulsed is very different from 5,000mw not pulsed.



The other important factor is the amount of energy that is transferred from the light source to the tissue. This is measured in Joules(J) and it is calculated by multiplying the power(mW) by time (seconds). 1 Watt for one second is equal to one Joule. If you apply 100 mW for 40 seconds this equates to 4 Joules. 75mW for 20 seconds is 1.5 Joules etc.

Multiply watts by seconds to get joules. A 1 Watt device consumes 1 Joule of energy every 1 second. If you multiply the number of watts by the number of seconds, you’ll end up with joules. To find out how much energy a 60W light bulb consumes in 120 seconds, simply multiply (60 watts) x (120 seconds) = 7200 Joules.Source

Reporting PBM Metrics Remains A Problem

There is a problem of standardisation of metrics within the literature on photobiomodulation. Some studies will give measurements in milliwatts while others will give measurements in milliwatts per centimetres squared. Compounding that confusion is the fact that sometimes the measurements in mW/cm2 are descriptive of a light beam which covers the entirety of that area and other times light beam cover only part of that area and the average power will be calculated over the total area. This is one of the reasons why you can see seemingly enormous variance in stated doses. (There are actual enormous differences in some instances too).

Sometimes it possible to calculate and standardise these doses for comparison across studies, sometimes there is not enough data to do that. So you need to look at a lot of studies to get a good idea of dosing in the real world. Lasers tend to have very small apertures, they cover a very small areas of tissue. For this reason it makes sense to state laser power in milliwatts and laser energy in Joules. LED light sources can cover much more area and so mW/cm2 and J/cm2 are more practical measurements for LEDs. (Generally LEDs cover centimeters squared areas, while lasers only illuminate tiny points)

Red and Near-Infrared Device Tests.

It is important to understand that the energy getting to the tissue will drop off pretty steeply as the light source moves away. This may be calculated by using the inverse square law (assuming you have an initial figure to work from). These calculations are extremely tricky in the real world. In reality it is better to measure the light power output or to buy from source you trust, one that lists the output at a number of different distances. I’ve also tested a redlightman device (670nm) and found it matched the output listed on their site. 

The other relevant factors are dosing frequency, how many times per day, many times per week, how many treatments per session, and how often a session needs to be repeated to maintain efficacy.

Inverse Square Law

All of the studies that have seen on thyroid tissue in animals and humans have been using laser sources. I’m thinking in terms of LED for this article. The reason for this is that they are cheaper and many people have them already. There are a number of reasonably cheap laser devices too (Laspot GDP-1 and B-Cure). Laser devices cover a much smaller area and application is much more involved. The usual laser procedure involves illuminating dozens of points marker with a mold as shown below. Tricky for someone to apply themselves at home.

Data From The Thyroid Studies On Low Level Laser Therapy


The human studies with positive outcomesthat did not use pulsing technology used power output of 50-70mW. Animal studies have shown positive effects as low as 30mW in one instance. Negative effects were seen in rabbits using a power output of 150mW.

From this you might conclude that 50-70mW can be safe and effective. At some point it seems there may be negative effects in humans, where that is relative to rabbits (150mW) is unclear. As we know there is effect in the 50-70mW range I don’t see any reason to go near 150mW. There may be a need for slightly more power with LED sources. Perhaps 50-100mW/cm2 with an LED would be a good start.


Most of the positive studies are using near infrared (830 and 890nm). The non pulsed studies are mostly using 830nm. Pulsing studies mostly use 890nm. There is positive data too regarding T3 and T4 using 780nm in mice. One rat study uses visible red (632nm) to protect T3, T4 and TSH levels from the anti-thyroid effects of ionizing radiation. So there is only positive data for 830nm (close to 820 peak) but no data showing other wavelengths are ineffective.

I suspect the peak wavelengths (620, 680, 760, 820) mentioned previously in the article would all have good effects on the thyroid but the strongest argument from the available data is for 830nm. As near infrared has deeper penetration it may require slightly more power or slightly more time when using visible red light. Slightly more power might be required using light further away from those peaks also.

Time and Energy

There is huge variability in reporting here, mostly in how the data is reported. The best data in humans (non-pulsed) states energy of 707J. But the area that the beam is focused on is only 0.2827mm2. This seems to work out to about 2 joules/cm2 averaged and that calculation is also mentioned in the study. I am inclined to think that 2J/cm2 is a better real world representation of what’s going on here than 707J. I also thing under-dosing is a better idea than overdosing, for reasons mentioned at the beginning of this article.

That study uses applications of 40 seconds per point (laser) at 50mW (830nm). There are a few dozen points used in this process which uses a clear mask/mold over the thyroid gland with holes used to identify the “points”. Points are illuminated for 40 seconds before moving on to the next. The fact that this laser device is a thin beam compared with LED and general recommendations in other treatments are often around 4J per point makes me suspect that slightly more than 2J/cm2 might be better when using LEDs. Maybe 3-5J/cm2 at 50-100mW using 830nm. If using red light or light further from the peaks then my guess would be closer to 5J/cm2 than 3J/cm2. Many leaps have to be made here because of the difference in laser and LED devices.


Positive animal studies are reported using 50mW with doses up to 20J/cm2, whereas the rabbit study showing thyroid inhibition is only 5J/cm2 but at 150mW vs 50mw. There may be more margin of error on the higher end with time and energy(J) than with power(mW).

Using LEDs To Treat Hypothyroidism?

Best Guess At Useful Treatment Parameters Using An LED Device

830nm at 75mW/cm2 for 60 seconds (4.5J/cm2) covering each side of the gland

100mW/cm2 for 50s (5J/cm2) 

Increase power or dose/time by 10-20% when moving away from that range using something around 620, 680 or 760 – except perhaps around 890nm where I think those measures might hold as penetration would be closer to that of 830nm.

The positive data is at 830nm, which is in the range of the 820nm peak. It might be wise to assume that light around these peaks would be more likely to work than light away from those peaks. The cost of a 620, 680, 760 or 820-830 range LED with rated power output is going to be close to a non-specific device at 650 or 850nm once you include a light meter of some sort. This is why I couldn’t bother with the infrared CCTV illuminator, they could be useful but must be tested for output. I found massive variances in output from those devices.

Treatment Frequency Of Light Therapy

Most experiments have used 10 – 15 sessions in the initial phase. One per day, 5 days a week for 2-3 weeks and repeated after 3 months. Benefits are seen after weeks and benefits peak after a few of months, after which when they begin a very slow decline. Some studies show 2nd and 3rd phases of 10 treatments a few months after the initial treatment build on improvements made in the first series of treatments. There doesn’t seem to be any reason to dose more frequently. A minority of people have almost complete recovery after one series of treatments.

The studies that use high (peak) power (pulsed) infrared on multiple parts of the body have shown that treating the thyroid gland is most effective for lowering TSH and increasing thyroid hormones, but treatments to the eardrum and clavicle area, along with irradiation of the blood vessels in the elbows are more efficient in correcting autoimmune processes. I’d assume similar dosing could be applied to the elbows and clavicle at least.

Vasal Tympanic Membrane

The first treatment was along the vasal tympanic membrane, or the eardrum. This was replicating a method developed in russia a decade earlier for use in rheumatoid arthritis. A number of the Russian and Ukrainian studies on light therapy treatment of hypothyroidism describe using light on areas other than the thyroid gland in order to correct autoimmune processes.

This particular method uses a fibre optic – transmitting the light from the laser device to the eardrum. The peak power output was 2 watts (2,000mW), far higher than the power used in non-pulsed studies (50-70mW). The beam wavelength was 890 nm, invisible – near infrared A (NIRA).

The pulse frequency was 3000 Hz. The duty cycle is not given, duty cycle is the percentage of time spent in the light-on stage versus light-off stage. I’ve read that where duty cycle is not given it can be assumed as 50-50. Treatment time was three minutes on each ear  for 10 days.

Thyroid – Thymus – Clavicle

The second treatment included irradiation of the thyroid gland itself. This treatment used the same 890 nm laser device with the same 3000 Hz pulse rate. The peak power in this treatment was 5 W (5,000mW).

The treatment time here was 64 seconds on each lobe of the thyroid, 64 seconds above the thymus gland and 32 seconds on the blood vessels of the supraclavicular fossa (region of the trapezius just above the clavicle).

This treatment was repeated for 10 days. The study does not say that treatments were conducted on consecutive days but it also doesn’t give any information on days between treatment, (my assumption is that it is 10 consecutive treatments).

Power And Distance

In order to dose properly you would need to find out at what distance your device emits mW in the range you want. Then you would need to create a setup whereby you can place the light at that distance from your thyroid and then time your exposure on both sides. I put my device on a high table or window ledge and place my solarmeter at the edge. I move my light back until I get the power output I want. Then it’s a matter of putting your gland where the meter was, using a mirror to see the light is hitting the right area, and using a stopwatch to time exposure.

There are many other ways you could do the same thing. If your device lists power output at distance then you might just need to measure distance. I’ve measured an redlightman 670 device and it matched the specs on that site. I’ve also measured a Laspot GDP-1 laser and found it be be perhaps 30% more powerful than specified. The infrared CCTV illuminators vary enormously in output, from 90mW at 0 inches to 200mW at 5 inches. These should not be used without being tested for power output first. See my previous articles testing some of these devices but do not assume a similar looking device will give you the same output.

Tests 1
Tests II

Do your own research!

“Like a good champagne It invigorates and stimulates; indulged in to excess, it intoxicates and poisons
~Sir Henry Gauvin – On Light Therapy~

Use the code RESONANTFM to get 10% off all redlightman LED devices.