Low-level laser in the treatment of patients with hypothyroidism induced by chronic autoimmune thyroiditis: a randomized, placebo-controlled clinical trial
- First Online:
- Cite this article as:
- Höfling, D.B., Chavantes, M.C., Juliano, A.G. et al. Lasers Med Sci (2013) 28: 743. doi:10.1007/s10103-012-1129-9
- 467 Views
Chronic autoimmune thyroiditis (CAT) is the most common cause of acquired hypothyroidism, which requires lifelong levothyroxine replacement therapy. Currently, no effective therapy is available for CAT. Thus, the objective of this study was to evaluate the efficacy of low-level laser therapy (LLLT) in patients with CAT-induced hypothyroidism by testing thyroid function, thyroid peroxidase antibodies (TPOAb), thyroglobulin antibodies (TgAb), and ultrasonographic echogenicity. A randomized, placebo-controlled trial with a 9-month follow-up was conducted from 2006 to 2009. Forty-three patients with a history of levothyroxine therapy for CAT-induced hypothyroidism were randomly assigned to receive either 10 sessions of LLLT (830 nm, output power of 50 mW, and fluence of 707 J/cm2; L group, n = 23) or 10 sessions of a placebo treatment (P group, n = 20). The levothyroxine was suspended 30 days after the LLLT or placebo procedures. Thyroid function was estimated by the levothyroxine dose required to achieve normal concentrations of T3, T4, free-T4 (fT4), and thyrotropin after 9 months of postlevothyroxine withdrawal. Autoimmunity was assessed by measuring the TPOAb and TgAb levels. A quantitative computerized echogenicity analysis was performed pre- and 30 days postintervention. The results showed a significant difference in the mean levothyroxine dose required to treat the hypothyroidism between the L group (38.59 ± 20.22 μg/day) and the P group (106.88 ± 22.90 μg/day, P < 0.001). Lower TPOAb (P = 0.043) and greater echogenicity (P < 0.001) were also noted in the L group. No TgAb difference was observed. These findings suggest that LLLT was effective at improving thyroid function, promoting reduced TPOAb-mediated autoimmunity and increasing thyroid echogenicity in patients with CAT hypothyroidism.
Chronic autoimmune thyroiditis (CAT) is an organ-specific autoimmune disease and is the most common cause of acquired hypothyroidism . The thyroid follicular cell injury in CAT patients is primarily mediated by T cells. Additionally, cytokines, chemokines, and thyroid autoantibodies contribute to an enhancement of the autoimmune response [2–5]. This process promotes the progressive destruction of thyroid follicle cells and significantly increases the risk of subsequent hypothyroidism . Once established, hypothyroidism requires a lifetime of levothyroxine (LT4) replacement in most patients.
Evidence suggests that both LT4 replacement and selenium supplementation may be associated with reduced concentrations of thyroid peroxidase antibodies (TPOAb) in CAT patients [7–9]. Despite a possible partial suppression of the autoimmune process, no improvements in thyroid function have been observed when such an approach is used with concurrent hypothyroidism. Currently, no successful treatments are available for the underlying CAT pathological processes.
Low-level laser therapy (LLLT) is a noninvasive, painless, low-risk, and low-cost method that uses the interaction of light with molecular structures to promote anti-inflammatory effects and the regeneration of biological tissues. LLLT has been shown to be an effective treatment for rheumatoid arthritis, a disease also caused by autoimmune mechanisms , and to aid in the regeneration of various injured tissues . In healthy animals, LLLT improves thyroid microcirculation [12,13] and elevates serum concentrations of T3 and T4 . In CAT patients, LLLT of the thymus, thyroid, and supraclavicular fossa (vascular irradiation) induces systemic immunomodulation .
Höfling et al.  have evaluated LLLT actions in patients with CAT hypothyroidism. Their study suggests that LLLT causes an improvement in thyroid function, a reduction in TPOAb, and an increase in echogenicity on thyroid ultrasonography. These results are encouraging, but no randomized clinical trials have been conducted to corroborate their preliminary results. Thus, the objective of this study was to evaluate the efficacy of LLLT in patients with CAT hypothyroidism through a randomized clinical trial that examined the following parameters: thyroid function, antithyroid antibodies, and ultrasonographic (US) criteria.
Materials and methods
The patients involved in this study were recruited from the Thyroid Outpatient Clinic of the Hospital das Clínicas, University of Sao Paulo Medical School (HC-FMUSP), and the study was conducted at the Radiology Institute of the HC-FMUSP. The thyroid patients were recruited from a population that is exposed to a sufficient intake of daily iodine .
Patients were eligible for the trial if they were between the ages of 20 and 60 years old, had been receiving stable doses of LT4 therapy for CAT hypothyroidism, and had normal (or near normal) serum concentrations of T3, T4, free-T4 (fT4), and thyrotropin (TSH). The appropriate LT4 replacement dose for each patient was determined by the doctors of the Thyroid Outpatient Clinic.
A CAT diagnosis was defined by the presence of the following criteria: (1) high serum concentrations of TPOAb and/or antithyroglobulin (TgAb) antibodies (>100 U/mL); (2) clinical and biochemical hypothyroidism that required LT4 replacement; and (3) a US pattern that was compatible with CAT [1,5,18].
The exclusion criteria included the following: (1) the use of immunosuppressants, immunostimulants, and drugs that interfere with the production, transport, and metabolism of thyroid hormones (e.g., corticosteroids, lithium, and amiodarone); (2) CAT with normal thyroid function; (3) CAT with subclinical hypothyroidism; (4) thyroid nodules; (5) hypothyroidism stemming from postpartum thyroiditis (up to 18 months after gestation); (6) a history of Graves’ disease; (7) prior treatment with radioiodine; (8) tracheal stenosis; (9) pregnancy; (10) a history of exposure to ionizing irradiation and/or neoplasia in the cervical area; (11) previous thyroid surgery; (12) thyroid hypoplasia; (13) ectopic thyroid; and (14) serious illness (e.g., cancer, ischemic coronary artery disease, stroke, kidney, and liver failure).
Of the 108 patients who initially presented with a high probability of inclusion, 47 were subsequently excluded because they did not fully meet the eligibility criteria, 14 lacked the time or transportation to get to the hospital, and 4 declined to participate. In the end, 43 patients were included for randomization.
HC-FMUSP’s Research Ethics Committee approved this study and the patient consent forms. All the patients signed the consent forms voluntarily.
The baseline characteristics of the trial groups
L group (n = 23)
P group (n = 20)
Mean age ± SD (years)
44.26 ± 9.76
42.1 ± 11.56
Mean disease duration ± SD (years)
4.61 ± 3.38
4.52 ± 3.47
Body mass ± SD (kg)
72 ± 6.2
72.7 ± 8.52
4 (17.39 %)
4 (20.0 %)
Diabetes mellitus type 2
1 (4.38 %)
1 (5.0 %)
5 (21.74 %)
4 (20.0 %)
8 (34.78 %)
7 (35.0 %)
1 (4.38 %)
1 (5.0 %)
For the study randomization, a computerized random-number generator was used to individually assign the patients to either the LLLT group (L) or the placebo group (P). The computer-generated randomization list contained the codes for the 60 interventions. The codes were deposited into opaque envelopes that were numbered 1–60 to indicate the group that each patient was allocated to. The envelopes were opened by the investigator responsible for the interventions (Höfling DB) only at the time of each patient’s first therapeutic procedure; this practice guaranteed that the investigator was unaware of the group to which each patient was assigned until the envelope was opened. The researcher (Höfling DB) needed to be aware of the patient assignments at the time of treatment because it was necessary to adjust the equipment to administer LLLT or placebo. Except for this investigator and the statistician, all of the study participants, including the patients, Thyroid Outpatient Clinic physicians, laboratory personnel, and US researcher, were blinded to the treatment assignments for the duration of the study.
Ultrasonography was used to mark the thyroid limits on the skin with a dermographic pen. Next, we created a mask (mold) for each patient using an adhesive plastic material that displayed the following demarcations: the gland contours (which contained 3-mm diameter perforations separated by 1 cm), the jugular notch of the sternum, and the prominence of the thyroid cartilage. The last two structures were easy to detect, which made placing the mask at the exact location of the thyroid simple and reproducible. The total number of perforations within the contours of the gland varied based on the size of the thyroid projection area on the skin, which was determined with the aid of ultrasound. Thus, the larger the demarcated area, the greater the number of irradiated points. The device tip was placed perpendicular to and in contact with the skin through the perforations located within the boundaries of the thyroid in the mold.
Both groups underwent 10 sessions of LLLT or placebo treatment twice a week for 5 weeks. The laser equipment was calibrated before each procedure.
The L group was treated with a continuous-wave diode laser device (830 nm, infrared) with a beam area of 0.002827 cm2 (Thera Lase™, DMC®, São Carlos, SP, Brazil) using the punctual method in continuous emission mode at an output power of 50 mW, a fluence of 707 J/cm2 (40 s at each point of application) and an irradiance of 17.68 W/cm2. Thus, the total energy deposited at each point was 2 J.
The P group was treated using the same method and equipment except that an ordinary nonlaser red light with an output power of 0.1 mW, a fluence of 1.41 J/cm2 and an irradiance of 0.03536 W/cm2 (the placebo), indistinguishable from the laser beam, was used. Therefore, the patients were blinded to which treatment they received.
We evaluated the presence of adverse effects from the procedures (LLLT and placebo) and medication use during the intervention and follow-up periods. The procedures and the patient follow-ups were performed in the Laser Medical Center and in the Thyroid Outpatient Clinic of the HC-FMUSP.
The primary outcome measure was improved thyroid function, as assessed by a significant reduction in the levothyroxine dose required to achieve normal T3, T4, fT4, and TSH concentrations after 9 months of follow-up. The secondary outcome measures were the serum concentrations of thyroid autoantibodies and the US parameters.
Criteria for reintroducing LT4 therapy
LT4 was discontinued 30 days after the last session of the LLLT or placebo treatment, just after the postintervention US. The LT4 therapy remained suspended in the patients with biochemical euthyroidism. The patients with subclinical hypothyroidism (TSH concentrations between 4.5 and 10 μU/mL and normal fT4) remained without LT4 for at least a 60-day follow-up period. The patients who exhibited biochemical hypothyroidism with TSH <50 μU/mL also remained without LT4 for a 60-day follow-up period to assess the need for LT4 therapy. LT4 was reintroduced 30 days after discontinuation to the patients who had biochemical hypothyroidism with TSH ≥50 μU/mL. LT4 was reintroduced in the second month of the follow-up period to the patients with biochemical hypothyroidism (TSH >10 μU/mL and fT4 <0.7 ng/dL) and on the sixth month of the follow-up period to patients with subclinical hypothyroidism. As planned, LT4 was reintroduced gradually until the patients reached the dose required to normalize their T3, T4, fT4, and TSH concentrations.
Venous blood samples were obtained by direct venous puncture in the morning, after an overnight fast, and before ingesting any medication. These samples were collected into tubes containing separator gel (with no anticoagulant) and identified with a barcode. These tubes were centrifuged at 3,000 rpm for 10 min and transferred to the AutoDELFIA® equipment (Wallac Oy PerkinElmer®, Turku, Finland), which sent the test results to the Interface System (LIS—Laboratory Systems Interface).
The characteristics of the assays used for the biochemical measurements
Intra-assay coefficients of variations/%
Interassay coefficients of variations/%
< 35 U/mL
< 35 U/mL
All of the US examinations were performed by a single blinded investigator.
The B-mode US and the real-time computerized grayscale histogram analysis (histogram) were performed using a Voluson 730 PRO™ device (General Electric Co®, Milwaukee, WI, USA) and the power Doppler US utilized an HDI-5000™ device (Philips Medical Systems®, Bothell, WA, USA). Both devices were attached to a broadband linear probe (5–12 MHz).
The ultrasound examinations were performed in patients who were administered the same preintervention and 30 days postintervention LT4 doses.
The B-Mode US examination was performed in the following manner. First, we estimated the thyroid volume  by adopting a reference value of 6–16 cm3 . Second, the texture was subjectively classified as either homogeneous or heterogeneous. Third, the echogenicity was evaluated using the histogram approach, which is an objective, quantitative, and reproducible method [21–24] that uses a grayscale ranging from black (value = 0) to white (value = 255). In each region of interest (ROI), a histogram of the mean ± 2 SD (standard deviations) grayscale values was plotted. The mean values plotted by the histogram originated from the upper, middle, and lower thirds of the longitudinal scan of the thyroid lobes and the transversal scan of the sternocleidomastoid muscles and prethyroid muscles. The echogenicity index (EI) was calculated as the ratio of the average histogram values of the thyroid gland and the values for the adjacent muscles (using the same brightness gain for both).
We calculated the sample size based on data from a pilot study . An improvement in thyroid function was defined to be a 50-μg reduction in the average necessary daily dose of LT4 for the treatment of patients. Using the PS Power and Sample Size Calculation® software (version 3.0, Dupont WD, Plummer WD, Vanderbilt Biostatistics Department, Nashville, TN, USA), an alpha of 5 %, a desired power of 80 %, and an SD of 45 μg in the daily average LT4 dose, we estimated that a total of 40 randomized patients were needed.
The statistical testing was primarily performed using the SPSS® software (version 16.0, SPSS Inc., IBM® Headquarters Company, Chicago, IL, USA). The results were compared by unpaired Student’s t tests for the normally distributed data and by Mann–Whitney tests for the data that departed substantially from a normal distribution. Fisher’s exact test was used to analyze the categorical data. The baseline clinical data are presented as the mean ± SD, and the outcome results are presented as the mean ± confidence interval (CI). Two-sided P < 0.05 were considered statistically significant.
A summary of the statistical analysis for the trial groups
n = 23, Mean
n = 20, Mean
n = 23, Mean
n = 20, Mean
(CI of 95 %)
(CI of 95 %)
(CI of 95 %)
(CI of 95 %)
LT4 dose (μg/day)
Thyroid volume (cm3)
Surround muscles histogram
All 43 randomized patients underwent 10 sessions of LLLT or placebo treatment, were followed for 9 months after the LT4 withdrawal, and were included in the final statistical analysis (Fig. 2). During this period, the patients did not ingest any medications that would interfere with the study results. No adverse effects (pain, heat, or other discomfort in the treated region) were observed in either group.
The primary outcome measure was thyroid function, as assessed by the LT4 dose needed to obtain normal mean concentrations of TSH, T3, T4, and fT4 in the ninth month of the follow-up period.
Additionally, 30 days after the LT4 suspension, the L-group patients showed TSH concentrations (27.41 ± 15.33 μU/ml) significantly lower than those of the P group (63.73 ± 28.91 μU/ml; P = 0.007; Fig. 3). At that time, significantly higher fT4 concentrations were also observed in the L group (0.72 ± 0.10 ng/dL) compared to the P group (0.55 ± 0.11 ng/dL; P = 0.022; Fig. 3).
A reduced LT4 dose or complete cessation of therapy was possible in 22 of the 23 patients (95.7 %) in the L group. In the P group, a reduced LT4 dose was observed in only 1 of the 20 patients (5 %). There was no need to reintroduce LT4 before the ninth month of the follow-up period in 11 of the 23 patients (47.8 %) in the L group, but all of the P-group patients required LT4 reintroduction. None of the L-group patients needed increased LT4, while 8 of the 20 patients (40 %) in the P group did.
Before the intervention, 23 of the 23 patients (100 %) in the L group showed high concentrations of TPOAb and 17 of the 23 (74 %) showed high concentrations of TgAb; in the P group, 18 the 20 (90 %) and 16 of the 20 patients (80 %) exhibited high levels of TPOAb and TgAb, respectively.
All of the patients from both groups exhibited undetectable TRAb levels prior to the interventions.
In the L group, the preintervention US showed that 16 of the 23 patients (66.6 %) had a normal thyroid volume, 6 of the 23 (26.1 %) had a goiter, and 1 of the 23 (4.35 %) had a reduced thyroid volume. In the P group, 12 of the 20 patients (60 %) had a normal thyroid volume, 5 of the 20 (25 %) had a goiter, and 3 of the 20 (15 %) had a reduced thyroid volume. After the LLLT, the thyroid volume was normalized in four of the six patients (66.6 %) and in the single (100 %) patient with a low thyroid volume.
A comparison of the observed frequencies of normal and abnormal postintervention US volumes in the L and P groups
In both groups, no alteration of the thyroid parenchymal texture was observed, either pre- or postintervention; it remained heterogeneous, and no evidence of nodules was noted.
An increased EI was observed in 22 of the 23 patients (95.7 %) in the L group and in only 3 of the 20 patients (15 %) in the P group (Table 3). The average EI increased significantly at 30 days post-LLLT compared to postplacebo treatment (Table 3, Fig. 4).
We tested the effectiveness of LLLT in patients who had CAT-induced hypothyroidism. We found that LLLT improved thyroid function, as measured by a reduction in the mean dose required for LT4 replacement therapy. In addition, we found that the TPOAb concentration was reduced and that the ultrasonographic echogenicity improved.
LLLT was introduced to clinical practice for its analgesic and anti-inflammatory properties and its ability to regenerate biological tissues [10,11,26]. In recent years, evidence has been published, suggesting that LLLT (with red to near-infrared light) operates at the molecular level in animal and human tissues by interacting with photoacceptors in the respiratory chain, probably the cytochrome c oxidase enzyme . This interaction serves to increase the production of ATP, reactive oxygen species, and nitric oxide while augmenting the intracellular Ca2+ levels. These initial events trigger a cascade of reactions that stimulate the signaling proteins responsible for increased expression of growth factors and cytokines that promote cell regeneration and protection [11,27,28].
In the thyroid tissue of healthy rats, LLLT stimulates microcirculation [12,13], and we speculate that this effect may have assisted in thyroid follicle cell regeneration. Furthermore, increased T3 and T4 concentrations have been found in healthy mice 7 days postinfrared laser radiation  without any microscopic evidence of injury of the thyroid follicles. Thus, it is plausible that LLLT can promote improved thyroid function.
As in a previous study , before withdrawing LT4, we opted to wait 30 days to ensure that there was sufficient time to detect any effect of LLLT on the regeneration of thyroid tissue. In the present study, a marked reduction in or even elimination of the LT4 requirement for at least 9 months was observed in the patients who underwent LLLT. In addition, an increased need for LT4 was detected in the P group during the follow-up period. This increase may have been associated with deterioration in thyroid function caused by the continuing CAT process.
We also found that 30 days after discontinuing LT4, the L-group patients presented with lower TSH and higher fT4 average concentrations than did those of the P group (Fig. 3). Because the patients were not on LT4 replacement at this time, the TSH and fT4 concentrations can be used as objective and direct indicators of thyroid function. Hence, these results suggest that improved thyroid function was already present on the 30th day of the follow-up period in the patients who received LLLT, which may be related to the regeneration of follicle cells.
In the ninth month of the follow-up period, we observed a reduction in the TPOAb, but not the TgAb, concentrations in the L group. High TPOAb and TgAb concentrations indicate the presence of thyroid autoimmunity , and their reduction suggests a modulation of the autoimmune response.
A red, or near-infrared, laser has been demonstrated to inhibit gene expression and/or reduce the serum concentrations of proinflammatory cytokines, such as tumor necrosis factor alpha (TNF-α), interleukin (IL)-1β, IL-2, IL-6, IL-8, and interferon gamma (IFN-γ). Furthermore, low-level laser stimulates the expression of regulatory cytokines, such as transforming growth factor beta (TGF-β) [11,29–33], thus promoting immunomodulation. Elevated levels of proinflammatory cytokines, such as IFN-γ, TNF-α, IL-2, and IL-6, and reduced TGF-β levels may play a role in CAT pathogenesis [3,34–36]. A positive correlation has been established between type 1 T helper cells (which produce IFN-γ and TNF-α) and high concentrations of TPOAb . Hence, the suppressive effects of LLLT on proinflammatory cytokines may explain the reduction in TPOAb levels.
The postintervention EI of the L group was significantly greater than that of the P group. Hypoechogenicity is the most important sonographic variable demonstrating the presence of thyroid autoimmunity [38,39] and/or the destruction of the thyroid follicle structure . The reflection of ultrasound waves results from the difference in acoustic impedance at the boundary between two distinct media. This reflection occurs in the normal thyroid follicular structure, which has a high content of colloid surrounded by follicular cells. Thyroid follicular lesions  and the presence of lymphocytic infiltration  promote scattering of the US waves, which reduces sound reflection and hypoechogenicity. Thus, the increase in echogenicity obtained after the LLLT suggests follicular regeneration and/or a reduction in lymphocytic infiltration.
The parenchymal texture remained heterogeneous, and no evidence of nodules was found in either group. However, thyroid texture is a subjective variable, and it is particularly difficult to distinguish subtle differences.
Most of the L-group experienced a normalization of thyroid volume, regardless of whether the volume was initially increased or decreased, possibly because LLLT has different biological actions in different cell types . It is plausible that in atrophic glands, follicle cell regenerative activities predominate and that in a goiter, LLLT causes a reduction of inflammatory infiltrates .
To guarantee the internal validity of the study, we chose to use several exclusion criteria. That decision inevitably compromised the generalization of the results to those conditions. Only one man was included in this study due to the higher prevalence of CAT in women  and the ineligibility of the other men assessed during the sample selection. This lack of male subjects may affect the generalization of the results to males. Nevertheless, there is no reason to believe that the effects of LLLT differ by sex. The male patient in our study received LLLT and remained free of LT4 after the ninth month of follow-up.
LLLT is a procedure that uses no ionizing radiation and, consequently, does not lead to neoplastic cell transformation . Even so, it would be prudent to accumulate more experience in applying this technique to thyroid tissue before treating patients with glandular nodules. Long-term follow-up of patients undergoing photocoagulation of benign nodules with high-power lasers would be informative. When a high-power laser is used to treat the interior of a thyroid nodule, the LLLT actions are probably present in the nodule periphery or even in the surrounding nodules. This phenomenon is currently called “X,” or the residual effect of the laser, and it occurs because the concentration of light energy decreases from the region where the laser beam contacts the tissue to the periphery of the nodule, where the effects of the low-level laser energy can be manifested . Indeed, monitoring patients undergoing percutaneous laser treatment of cold benign thyroid nodules has revealed that high-power lasers are safe for thyroid treatment . It is important to note that none of the patients included in our study had nodules.
In summary, our results suggest that LLLT is effective at enhancing thyroid function; it reduced the organ-specific autoimmune responses mediated by TPOAb and increased the echogenicity of the thyroid. This is the first trial conducted to evaluate the efficacy of LLLT in patients with CAT-induced hypothyroidism. The results were consistent with those reported in our pilot study  and encourage new studies evaluating the use of LLLT in CAT patients to verify our findings. It should also be noted that this study followed the patients for 9 months; therefore, it will be important to assess the duration of the beneficial LLLT results. It is likely that the observed effects will not last indefinitely because CAT is a chronic disease; therefore, we expect that additional applications may be necessary.
The use of LLLT could be especially interesting in the early phases of CAT, when subclinical hypothyroidism is present. Perhaps, using this approach at an early stage can reduce the incidence of overt hypothyroidism. Further studies are needed to test this hypothesis, which may bring important implications for public health using a noninvasive, painless, low-risk, and low-cost procedure that is easily implemented in healthcare centers.
We would like to thank Berenice B. Mendonça, Suemi Marui, Noedir A. G. Stolf, Mikiya Muramatsu, Rosangela Itri, and Claudio Leone for their assistance and support in this study. This research project was supported by grants from the “Fundação de Amparo à Pesquisa do Estado de São Paulo” (FAPESP), “Conselho Nacional de Desenvolvimento Científico e Tecnológico” (CNPq), and a fellowship from “Coordenação de Aperfeiçoamento de Pessoal de Nível Superior” (CAPES).
Disclosure of proprietary interests
“I certify that I have no affiliation with or financial involvement in any organization or entity with a direct financial interest in the subject matter or materials discussed in the manuscript.”