Potential Risks and Benefits of Desiccated Thyroid Extract for the Treatment of Hypothyroidism: A Systematic Review

Abstract

Background:

Desiccated thyroid extract (DTE) is no longer recommended for the treatment of hypothyroidism but is still in use. This review aimed to summarize the available literature on treatment with DTE in adult patients with hypothyroidism.

Methods:

The search was conducted up until January 6, 2024, in six electronic databases. Two reviewers independently screened all the search results. The retrieved studies compared DTE treatment with levothyroxine or combination therapy with liothyronine and levothyroxine. The primary outcome was quality of life (QoL), and the secondary outcomes included symptoms, treatment preference, adverse effects, thyroid hormone levels, thyroid autoantibodies, cardiovascular measures, and gene polymorphisms in deiodinase enzymes.

Results:

In the qualitative synthesis, we included nine nonrandomized studies of interventions (NRSIs), two randomized clinical trials (RCTs), and three case reports. The overall quality of evidence was moderate to very low for the various outcomes. The RCTs found no difference between treatments regarding QoL and symptom score assessments. In the NRSIs, symptom and QoL assessments were in favor of DTE. The included studies indicated that DTE may cause an increase in heart rate, lower body weight, and lower high-density lipoprotein compared with other treatment regimens, but results were conflicting.

Conclusions:

Most studies of DTE treatment are hampered by an inferior design, and data on long-term effects and side effects are lacking. Two RCTs could not demonstrate any difference in QoL or symptom scores when comparing DTE with other thyroid hormone substitutions. Future trials of DTE in patients with hypothyroidism should be based on adequate study designs, validated measures of QoL, patients with reduced QoL, and the assessment of biomarkers reflecting long-term adverse effects.

Introduction

Desiccated thyroid extract (DTE), derived from the porcine or bovine thyroid gland, was the standard therapy for hypothyroidism until the mid-1970s,¹ but is no longer recommended by guidelines worldwide.² The primary concern of DTE is the high content of triiodothyronine (T3), compared with the human thyroid, which potentially leads to transient thyrotoxicosis. This may harm bone tissue and the cardiovascular system.^3

–7 A possible association between a high T3 level and breast cancer has also been described, but the evidence is conflicting.^8
–10

Even though levothyroxine (LT4) was introduced in 1949, it took some time before it became the first choice of therapy for hypothyroidism.¹¹ Despite a normal thyrotropin (TSH) level on LT4 treatment, 10–15% of the patients have impaired quality of life (QoL) when compared with the background population.^12
–14 It is debated whether this is caused by insufficient amounts of T3 in the tissues due to an inadequate conversion of thyroxine (T4) by deiodinases.¹⁵

Despite concerns about drugs containing T3, many patients request DTE treatment, incited by ongoing debates on social media.¹⁶ Even though there has been renewed interest in DTE, only 0–5% of the European endocrinologists prescribe DTE to their patients.^17

–22 Still, there are exceptions, as 17% of Swedish and 13% of Irish endocrinologists are more prone to prescribe DTE.^23,24 In this systematic review, we aimed to summarize the possible benefits and adverse effects of DTE treatment.

Materials and Methods

The study protocol, developed in line with the PRISMA-P statement,²⁵ was registered in the International Prospective Register of Systematic Reviews (PROSPERO), and was published on July 5, 2020 (ID: CRD42020185456).²⁶

Eligibility criteria

We wanted to identify available literature concerning the association between treatment with DTE in patients with hypothyroidism of age ≥18 years and various outcomes of interest. The primary outcome was the overall QoL, self-rated or assessed by a questionnaire instrument. Secondary outcomes were cognitive function, anthropometric measures, level of fatigue, thyroid hormone levels, adverse effects or events, patient preference for treatment, surrogate markers (e.g., biomarkers of bone metabolism and the cardiovascular system), interactions with gene polymorphisms, or any other relevant outcome reported in the related studies. The comparator was other thyroid hormone substitutions, placebo, or no treatment.

The TSH radioimmunoassay was developed in the early 1970s. Until then, various methods were used to diagnose hypothyroidism, including evaluation of symptoms, serum protein–bound iodine levels, and basal metabolic rate.²⁷ These methods are much less reliable than the assessment of TSH for diagnosing and monitoring patients with hypothyroidism.^27,28 Therefore, an additional eligibility criterion was added during the screening process, in that studies had to employ TSH in the clinical management of the patients. This led to the exclusion of 24 studies/case reports conducted >50 years ago (Supplementary Data S2: categorized as “Wrong method for diagnosis” in reason for exclusion).²⁷

Search

The search was conducted up until January 6, 2024, by the first author K.R.R. in the electronic databases CINAHL (EBSCO), Scopus, “Cochrane central register of controlled trials” (CENTRAL), and “Cochrane Database of Systematic Reviews” (CDSR) via Cochrane library, Embase Classic + Embase via Ovid, MEDLINE, and Epub Ahead of Print and In-Process, and other nonindexed citations via Ovid, without time restrictions. The search strategy was developed in collaboration with a research librarian at the University Library of Southern Denmark. We used medical subject headings (MeSH), CINAHL subject headings, Embase subject headings (Emtree), and text words related to our eligibility criteria. The complete search strategy is compiled in Supplementary Data S1.

Reference lists of reviews, guidelines, and meta-analyses found in the full-text screening were hand-searched for relevant references. A citation search from the included studies was conducted in Scopus. OpenGrey.eu and Clinicaltrials.gov were searched for unpublished literature.

Study selection

All records from the search were assessed and organized as references in the open-access software program Mendeley Desktop (version 1.19.8, Mendeley Ltd., London, United Kingdom). Duplicates were removed, and the remaining references were imported to the Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia; Available at www.covidence.org). Covidence removed additional duplicates and was used independently by two authors (K.R.R. and C.B.L.) for the title and abstract screening according to the eligibility criteria. The full text of potentially eligible references was retrieved, and both authors read and evaluated the articles. Any disagreement was discussed and solved by consensus with the contribution from author S.J.B., who had the deciding vote.

Randomized controlled trials (RCTs), prospective or retrospective observational studies (cohort and case–control), cross-sectional or longitudinal studies, case reports, and case series were all included. Qualitative, quantitative, and mixed-methods research were included. Nonoriginal studies, including editorials, letters to the editor, textbook chapters, and guidelines, were excluded. Only publications in English were included. The authors of conference abstracts were requested a full report, which was included if it met the eligibility criteria. The conference abstract was excluded if a full study report was not accessible or the author did not respond.

Data collection

For all included studies, characteristics (e.g., year of publication, sample size, study design, setting, intervention, outcome, and conflict of interest), demographic data of the participants (e.g., age, sex, and etiology of hypothyroidism), and results were extracted and compiled in tables. Data extraction was done twice by the first author. Results are presented as numbers and percentages for categorical data and mean with standard deviations or median with range for continuous data, as appropriate. Some results are presented narratively.

Quality assessment

The risk of bias for each study was evaluated by two authors (K.R.R. and C.B.L.). Judgment between the authors was cross-checked, and a consensus was reached if there were any disagreements.

The methodological quality of each nonrandomized study of intervention (NRSI) was assessed using the risk-of-bias assessment tool “Risk Of Bias In Non-randomized Studies of Interventions” (ROBINS-I).²⁹ The risk-of-bias assessment tool revised Cochrane risk-of-bias tool for randomized trials (RoB 2.0) was used for RCTs.³⁰ The authors were contacted if a study protocol or statistical analysis plan was unavailable. The overall quality of evidence was assessed using the “Grading of Recommendation, Assessment, Development, and Evaluation” (GRADE) approach.³¹

Meta-analysis

A meta-analysis of the included RCTs was conducted using the inverse-variance approach with fixed effects. Statistical heterogeneity was evaluated with visual inspection of the forest plots, chi-square tests, and I ²-statistics. The meta-analyses were performed using Stata software (V17.0; StataCorp, College Station, TX, USA). Because fewer than ten studies complied with the inclusion criteria for each outcome, we did not conduct a funnel plot to assess small-study effects.

Results

Literature search

Figure 1 illustrates the literature search and study selection process. The full-text screening led to the exclusion of 146 articles (Supplementary Data S2). In total, fourteen articles met the inclusion criteria, here among two RCTs, nine NRSIs, and three case reports.^{32

–45}

FIG. 1.

PRISMA flow diagram illustrating the search and screening process for the review “potential risks and benefits of desiccated thyroid extract for the treatment of hypothyroidism: A systematic review.” ¹Mendeley Desktop (version 1.19.4, Mendeley Ltd., London, United Kingdom). ²Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia).

Study characteristics

The detailed characteristics of the included studies and baseline descriptions of the study participants are compiled in Tables 1 and 2, respectively. In six studies, DTE was compared with LT4 monotherapy,^{32,34,35,43
–45} while one study compared DTE with either LT4 monotherapy or no treatment.³⁶ The remaining studies (case reports not considered) compared three treatment regimens, that is, DTE, LT4 monotherapy, and LT4 and liothyronine (LT3) combination therapy.^33,37
–39 Most studies used Armour thyroid tablets, one study used both Armour and ERFA, and one study used Thyranon (Table 2).^{32

–36,38,43,45} Three studies did not describe the type of DTE used, besides that it was “prescribed” DTE (Table 2).^37,39,44

Table 1.

Study Characteristics

1^st author, year of publication, country	Study design	Sample size (n)	Intervention/comparator	Follow-up	Outcomes of interest	Conflict of interest/funding
Baisier, 2001,³⁶ Belgium	Retrospective before-and-after	318^a	DTE vs. LT4 or untreated	NI	Symptom score, 24-hour urine fT3	NI/NI
Bindels, 1988,³⁵ The Netherlands	Retrospective before-and-after	35	DTE vs. LT4	DTE: mean of 90 months (range: 6–120 months)	TT3, TT4, TSH	NI/NI
				LT4: mean of 24 months (range: 2–49 months)
Heald, 2021,³⁵ UK	Consecutive case-series	31^b	DTE vs. LT4	Range: 6 months—2.2 years	ThyPRO-39, EuroQoL EQ-5D-5L, EQ-5D-5L utility score, EQ-VAS, fT4, fT3, TSH, BMI	None/NI
Hoang, 2013,³² USA	Randomized double-blinded crossover	70^c	DTE vs. LT4	Two periods of 16 weeks^d	TSQ-36, GHQ-12, WMS-IV, BDI, body weight, heart rate, blood pressure, lipid profile, TSH, fT4, TT4, TT3, treatment preference	None/Funded by Walter Reed National Military Medical Center Institutional Review Board
Kuye, 2020,⁴⁴ USA	Retrospective matched cohort	870^e	DTE vs. LT4	Three years	Percent of in-range TSH, visit-to-visit TSH-variability, percent of patients with TSH in range during follow-up	None/NI
Lee, 1998,⁴³ USA	Prospective cohort	13^f	DTE vs. LT4	One year	TSH, TT4, fT4 index, TgAb, TPOAb	Forest Pharmaceuticals provided DTE
Penna, 2023,³⁹ USA	Multiyear cross-sectional	5437^g	DTE vs. LT4 vs. LT4/LT3	Median treatment time: LT4: 6 years (IQR 1–16), DTE: 3 years (IQR 1–9), LT4/LT3: 3 years (IQR 1–9.5)	Health care utilization	One author is a consultant for AbbVie, Acella, Sention, Synthonics, and Thyron.Another author is a consultant for Knight, Ipsen, and Bayer. The other authors have no relevant disclosures/ funded by the National Institute of Diabetes and Digestive and Kidney Diseases, and the Diabetes Research and Training Center
Pepper, 2014,⁴⁵ USA	Retrospective cohort	150^h	DTE vs. LT4	Minimum 4 weeks (mean 2 years and 7 months)	Five-point Satisfaction Rating Scale, adverse events, TSH, TT3, TT4, body weight	None/no funding
Shakir, 2021,³³ USA	Randomized, double-blinded crossover	75ⁱ	DTE vs. LT4 vs. LT4/LT3	22 weeks (for each intervention)	TSQ-36, GHQ-12, WMS-IV, BDI, body weight, heart rate, blood pressure, lipid profile, TSH, TT3, TT4, fT4, treatment preference, deiodinase polymorphism	Consultancy for Allergan, a pharmaceutical company producing DTE/funding by Walter Reed National Military Medical
						Center Institutional Research Board
Surks, 1972,³⁸ USA	Observational before-and-after	3	DTE	1–24 hours after ingestion of DTE	T3, T4	NI/supported by US public health service and US Army contract
Tariq, 2018,³⁷ USA	Observational retrospective chart review/cross-sectional	94^j	DTE and LT4/LT3 vs. LT4	Mean: 27 months (median 22, range: 1–111 months)	Clinical signs and symptoms, TSH, fT4, fT3, adverse events, SF-20, treatment preference	None/NI

89 participants were LT4 treated and 832 were untreated at baseline, of whom 40 and 278, respectively, had follow-up.

35 patients were offered DTE, of whom 4 refused and 3 discontinued DTE.

78 were enrolled, 70 completed the study.

Medication was adjusted after six weeks aiming at TSH in the range of 0.5–3.0 μIU/L; hereafter patients continued the medication for an additional 12 weeks.

435 DTE treated and 435 LT4 treated.

Seven received DTE and 6 received LT4.

5106 received LT4, 252 received DTE, and 79 received LT4/LT3 combination therapy.

450 undergoing LT4 treatment were screened, 154 were switched to DTE because of failure to achieve clinical euthyroidism. Four patients were excluded from the TSH, TT4, and TT3 analyses due to post-thyroid cancer therapy or suppressive treatment for thyroid nodular disease.

90 were enrolled, 75 completed the study.

100 patients included initially, 60 patients received DTE (3 were excluded), and 40 received LT4/LT3 combination therapy (3 were excluded); 51 patients receiving DTE completed the cross-sectional study with SF-20.

BDI, beck depression inventory; BMI, body mass index; DTE, desiccated thyroid extract; EuroQoL EQ-5D-5L, european quality of life 5 dimension 5-level; EQ-VAS, european quality of life visual analog scale; fT3, free triiodothyronine; fT4, free thyroxine; GHQ-12, general health questionnaire; IQR, interquartile range; LT3, liothyronine; LT4, levothyroxine; NI, no information; SF-20, medical outcomes study short form-20 questionnaire; TgAb, thyroglobulin antibodies; ThyPRO-39, thyroid-related patient-reported outcome; TPOAb, thyroid peroxidase antibody; TSH, thyrotropin; TSQ-36, thyroid symptom questionnaire; TT4, total T4; TT3, total T3; T4, thyroxine; T3, triiodothyronine; WMS-IV, wechsler memory scale fourth version.

Table 2.

Baseline Demographic Characteristics of the Participants and Thyroid Supplement Dose in the Included Nonrandomized and Randomized Clinical Trials

1^st author	Age	Sex	Ethnicity (%)	Etiology of hypothyroidism (n or %)	Dose of thyroid supplementation at baseline (Mean ± SD or range)	Dose of desiccated thyroid extract (Mean ± SD or range)	Type of desiccated thyroid extract
1^st author	Years (Mean ± SD or range)	Female/male(n or %)	Ethnicity (%)	Etiology of hypothyroidism (n or %)		Dose of desiccated thyroid extract (Mean ± SD or range)
Baisier	Group 1 (LT4-treated at baseline)^a: 43.30 ± 11.02	Group 1: 36/4	Group 1: 100% Caucasian	NI	Group 1: LT4: 99.7 µg/day	Group 1: 233 mg/day		Armour thyroid tablets
	Group 2 (untreated at baseline)^a: 43.08 ± 12.09	Group 2: 228/50	Group 2: NI			Group 2: 200 mg/dayDosing: NI
Bindels	56.3 (21–87)	28/7	NI	AIT: 8	NI	50–100 mg/day	Thyranon thyroid tablets
				Thyroidectomy: 15^b		Dosing: NI
				RAI: 4
				Other/unknown: 8^c
Heald	49.4 ± 12.0 (26–77)	28/3	NI	AIT: 28Thyroidectomy: 2^d RAI: 1	NI	123.5 ± 6.5 mg/day (60–180 mg/day)Dosing: once daily (twice daily for two participants)	ERFA thyroid or Armour thyroid tablets

Hoang	50.7 (23–65)	53/17	Caucasian: 74.29%	AIT: 35	LT4: 112.4 ± 36.3 µg/day	80.63 ± 29.97 (43–172) mg/day^f	Armour thyroid tablets
			African American: 15.71%	Idiopathic: 14	(75–211 µg/day)^e	Dosing: once daily
			Asian: 5.71%	RAI: 10
			Hispanic: 4.29%	Postsurgical: 8
				Postradiation: 3
Kuye	63.4 ± 12.68^g	784/86	Non-Hispanic white: 83.4%	Primary hypothyroidism^h	NI	NI	NI
			Non-Hispanic black: 1.1%
			Hispanic: 5.1%
			Other: 10.3%
Lee	DTE: 50.1LT4: 52.8	NI	NI	AIT: 8RAI: 5	LT4: 525–1225 µg/week	NI	Armour thyroid tablets

Penna	LT4: 61.4 ± 16.0	LT4: 76.2% females	LT4: white (88.4%), black (5.3%), native American/ Alaskan (0.6%), Asian/ Pacific Islander (4.1%), multiple races (1.6%), Hispanic (8.8%)	NI	NI	NI	NI
	LT4/LT3: 60.7 ± 12.5	LT4/LT3: 97.1% females	LT4/LT3: white (95.7%), black (2.9%), native American/Alaskan (0%), Asian/Pacific Islander (0.7%), multiple races (0.7%), Hispanic (0.8%)
	DTE: 56.0 ± 14.0	DTE: 92.2% females	DTE: white (91.9%), black (3%), native American/Alaskan (0.6%), Asian/Pacific Islander (0.8%), multiple races (3.7%), Hispanic (7.7%)
Pepper	53.5 ± 12.5 (24–86)	139/15	NI	AITThyroidectomyRAIOtherⁱ	NI	92.3 ± 28.6 mgDosing: once daily	Armour thyroid tablets
Shakir	50 (29–65)	58/17	Caucasian: 77.3% African American: 16.0% Asian: 3.0% Hispanic: 2.7%	AIT: 46Idiopathic: 9RAI: 4Thyroidectomy: 16	LT4 (n = 64): 112.5 ± 28.7 µgDTE (n = 8): 71.2 ± 10.9 mgLT4/LT3 (n = 3): 97.3 ± 16.2 µg/5 ± 0 µg	Start of the study period: 76.7 ± 16.3 mg/dayEnd of the study period: 77.3 ± 16.8 mg/dayDosing: once daily	Armour Thyroid tablets
Surks	NI	NI	NI	NI	NI	60–120 mg/day	Armour Thyroid tablets
						Dosing: once
Tariq	54 (20–81)	DTE-group: 57/3	NI	AIT: 52%	NI	30 mg/day^j	NI
				Postsurgical: 22%		Dosing: NI
				RAI: 10%
				Other: 16%

Participants were divided into group 1 (receiving LT4 at baseline), 40 patients, and group 2 (untreated at baseline), 278 patients; in all, 318 patients received DTE (content: 23 µg triiodothyronine, 100 µg thyroxine).

Twelve patients had subtotal thyroidectomy due to thyrotoxicosis, nontoxic goiter, or AIT, and three had a total thyroidectomy due to thyroid cancer.

Four patients had unknown etiology, one had Schmidt’s syndrome, and three had congenital hypothyroidism.

Total thyroidectomy due to toxic goiter.

Two patients were taking DTE at inclusion (75 and 100 mg/day, respectively).

DTE content: 65 mg provides 38 µg/day thyroxine and 9 µg/day triiodothyronine.

The DTE and LT4 groups were matched on age, sex, and ethnicity.

Exclusion criteria: previous radioiodine therapy, panhypopituitarism, pregnancy, previous thyroid cancer, Graves’ disease, autoimmune thyroiditis, and prescription of more than one thyroid supplement.

No information about number of patients in each treatment category.

Content of T4:T3 ratio is 4:1.

AIT, autoimmune thyroiditis; DTE, desiccated thyroid extract; LT4, levothyroxine; LT3, liothyronine; n, number; NI, no information; RAI, radioactive iodine; SD, standard deviation.

Four studies evaluated QoL using various tools.^32,33,35,37 Other outcomes included symptoms of hypothyroidism, symptoms of depression, thyroid function measures, treatment preference, body weight, adverse effects, lipid profile, heart rate, and blood pressure. Gene variants in relation to symptoms and the response to treatment were investigated in one study.³³

Risk of bias

One NRSI was classified with moderate risk of bias with ROBINS-I, four NRSIs were classified with serious risk of bias, and the remaining four with critical risk of bias (Fig. 2, Supplementary Data S3). Overall, the RCTs were classified with some concern due to a missing protocol or an a priori statistical analysis plan (Fig. 2, Supplementary Data S4).

FIG. 2.

Risk-of-bias assessment in the nonrandomized trials using ROBINS-I (Risk Of Bias In Non-randomized Studies of Interventions) and in the randomized controlled trials using RoB 2.0 (Revised Cochrane risk-of-bias tool for randomized trials). The detailed evaluation with answers to all signaling questions is compiled in Supplementary Data S3 and S4. In RoB 2.0, Domain 5 is judged as “some concerns” in both the randomized controlled trials, as no a priori statistical analysis plan or protocol was available for either study. BH, bone health; BMI, body mass index; CVD, cardiovascular; EuroQoL EQ-5D-5L, European Quality of life 5-dimension 5-level; fT3, free triiodothyronine; fT4, free thyroxine; SF-20, Medical outcomes study short form-20 questionnaire; T3, triiodothyronine; T4, thyroxine; TSH, thyrotropin; ThyPRO, Thyroid-related Patient-Reported Outcome.

Quality of evidence

Including both RCTs and NRSIs in the certainty of evidence assessment, the overall rating was low for the QoL outcome (Supplementary Data S5). Based on the RCTs only, the body of evidence was higher. The certainty of evidence was low also for the secondary outcomes, such as symptom scores and thyroid hormone levels. The certainty of evidence for adverse effects was rated as very low due to the low number of cases and moderate for the cardiovascular surrogate markers.

Quality of life

In the two RCTs, QoL was evaluated with the General Health Questionnaire (GHQ-12), and no difference could be detected between the treatment regimens (Table 3).^32,33 In a fixed-effects meta-analysis, the overall Hedges’ g of GHQ-12 was −0.014 (95% CI: −0.37, 0.09; p = 0.23) when comparing DTE with LT4 treatment (Supplementary Data S6). The heterogeneity was estimated to be low (I ² = 0%).

Table 3.

Summary Table of the Associations Between Treatment with Desiccated Thyroid Extract and Quality of Life, General Health, and Symptom Scores

1^st author	Method	Quality of life, general health, and symptom scores
		Mean ± standard deviation (SD) or confidence interval (CI)
		Baseline	After intervention (DTE)	p value
Baisier⁴⁶	Symptom score^a:	Group 1 (LT4 at baseline) (n = 40): 10.72 ± 3.17 points	Group 1 (n = 40): 3.6 ± 2.6 points	Not available
	The maximum score is 16.A lower score indicates fewer symptoms	Group 2 (untreated at baseline) (n = 278): 10.1 points (SD not available)	Group 2 (n = 278): 3.6 points (SD not available)
Heald⁴⁷	EuroQoL EQ-5D-5L domains^b	Improvement from baseline to six months after DTE initiation across all domains (n = 31)		Not available
	EQ-5D-5L utility score^c	0.214 ± 0.338 (n = 31)	0.606 ± 0.248 (n = 31)	p < 0.001
	EQ-VAS^d	33.4 ± 17.2 (n = 31)	71.1 ± 17.5 (n = 31)	p < 0.001
	ThyPRO-39^e Composite score	Composite score (n = 29): 68.3 (60.9; 75.7)	Composite score (n = 29): 25.2 (18.7; 31.7)	p < 0.001
	Domain scores	ThyPRO-39 scores improved in all domains (up to several multiples of minimal important change)		Not available
Hoang⁴¹	GHQ-12, TSQ-36^f, BDI^g, and WMS-IV^h.	No difference in the TSQ-36, GHQ-12, BDI, or WMS-IV tests between treatments		NS
Shakir⁴²	GHQ-12, TSQ-36^f, BDI^g, and WMS-IV^h.	No difference in the TSQ-36, GHQ-12, BDI, or WMS-IV tests between treatments		NS
Tariq⁴⁸	SF-20ⁱ questionnaire	DTE-group (cross-sectional, no comparator) (n = 51):
		Question 1: 92% were feeling “excellent, very good, or good” about self-health.
		Question 2: 80–100% were not limited at all in their daily activities (eating, dressing, bathing, carrying groceries, climbing stairs, running, walking, strenuous sports, housework).
		Question 6: 86.8% denied having any limitations on social activities.
		Question 7: 76.5% denied anxiety.
		Question 8: 86.8% reported “feeling calm and peaceful.”
		Question 9: 88.2% denied feeling downhearted.
		Question 10: 88.2% reported “being a happy person.”
		Question 11: 100% denied any sense of hopelessness.
		Question 12: 10.3%: I am somewhat ill and 82.4%: I am as healthy as anybody I know.
	Question about “feeling better” on DTE versus LT4-monotherapy	92.7% claimed to have improved quality of life by switching from LT4 to DTE

Evaluates eight symptoms with a score from 0 to 2; 0 is no symptoms and 2 is full presence of symptoms.

Five dimensions of health: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Five levels can be chosen within each health dimension: no problems to extreme problems.

Utility scores: representing a preference-weighted measure of patients’ health evaluation from the EQ-5D-5L dimensions, a lower score signifies poorer perceived health.

EQ-VAS scale: self-rated score, 100 is “the best imaginable health state” and 0 represents “the worst imaginable health state.”

^eT

hyPRO-39 questionnaire, self-administered and measures QoL within 39 items (scores of 0–100); increasing scores indicate decreasing QoL.

Self-developed TSQ: 12 questions, higher scores indicate more symptoms.

BDI: self-rating scale of 21 items, scores of ≤10 indicate normal mood variation, scores of ≥11 indicate symptoms of depression, and scores of >20 reflect clinical depression.

WMS-IV includes Auditory Memory Index (AMI), Visual Memory Index (VMI), Visual Working Memory Index (VWMI), Immediate Memory Index (IMI), and Delayed Memory Index (DMI).

SF-20 is available at https://www.rand.org/health-care/surveys_tools/mos/20-item-short-form/survey-instrument.html. SF-20 was not tested in patients receiving LT4, and there is no preintervention data for comparison.

BDI: beck depression inventory; DTE, desiccated thyroid extract; EuroQoL EQ-5D-5L, european quality of life 5-dimension 5-level; EQ-VAS, european quality of life visual analog scale; GHQ-12, general health quality of life questionnaire; LT4, levothyroxine; n, number; ns, nonsignificant; QoL, quality of life; SF-20, medical outcomes study short form-20 questionnaire; ThyPRO, thyroid-related patient-reported outcome; TSQ-36, thyroid symptom questionnaire; WMS-IV: wechsler memory scale fourth version.

In consecutive case series, Heald et al. evaluated QoL with both the Thyroid-related Patient-Reported Outcome (ThyPRO-39) and the European Quality of Life 5-Dimension 5-Level questionnaire (EuroQoL EQ-5D-5L).³⁵ Both tools showed an improvement in QoL from baseline (treatment with LT4) to six months after initiating DTE (Table 3). For ThyPRO-39, all domains and the composite score improved significantly on DTE treatment. The same pattern was seen in the EuroQoL EQ-5D-5L questionnaire (Table 3).³⁵

Tariq et al. evaluated QoL with the Short Form-20 questionnaire (SF-20) (Table 3).³⁷ Most patients (92%) indicated that their self-health was excellent, very good, or good after initiating DTE compared with LT4. In addition, most of the participants (92.7%) claimed to have improved QoL after a switch of treatment to DTE.³⁷

Symptom score

In the study by Baisier et al., a self-developed hypothyroid symptom score tool was applied.³⁶ Patients unresponsive to LT4 or being treatment-naïve were all treated with DTE. The study did not compare the symptom score before and after treatment with DTE, but, nevertheless, the authors concluded that the mean symptom score declined in both groups following DTE treatment (Table 3). The primary outcome of the study by Tariq et al. was clinical signs and symptoms of hypothyroidism following DTE treatment, but these results were, for unknown reasons, not reported in the article.³⁷

In the two-armed RCT by Hoang et al., symptoms were evaluated with TSQ-36 (Thyroid-Symptom Questionnaire), and there were no differences between DTE and LT4 monotherapy.³² The three-armed RCT by Shakir et al. also used TSQ-36 and found no difference in symptoms between DTE, LT4, and LT4/LT3 (Table 3).³³ In addition, the Beck Depression Inventory (BDI) and Wechsler Memory Scale (WMS) were evaluated in these two RCTs, without finding any difference between treatments (Table 3). In fixed-effects meta-analyses of TSQ-36, BDI, and Wechsler’s memory indices, there were no differences comparing DTE and LT4 treatment (Supplementary Data S6), and the heterogeneity was low in all outcomes (I ² = 0%).

Thyroid hormone levels

Three studies found no difference in TSH between DTE and other treatment regimens,^34,43,45 two studies found a significantly higher TSH,^32,33 and one found a significantly lower TSH on DTE compared with LT4 treatment (Table 4).⁴⁴ One study found DTE superior to LT4 in maintaining plasma TSH within the reference range.³⁷

Table 4.

Association Between the Treatment of Hypothyroidism with Desiccated Thyroid Extract and Thyroid Hormone Levels

1^st author	Timing of blood samples	Thyrotropin (TSH) mean/geometric mean ± standard deviation (SD)	Triiodothyronine (T3) mean ± SD (or range)	Thyroxine (T4) mean ± SD
Baisier			24-hour urine fT3^a was higher when treated with DTE than LT4 or treatment-naïve patients (p value not given).
Bindels	No information.	Serum TSH^b was not different between DTE and LT4 regimens.	T3^b was higher on DTE than on LT4 ( p < 0.001): 2.9 ± 1.1 nmol/L vs. 1.9 ± 0.5 nmol/L, respectively.	T4^b was lower on DTE than on LT4 ( p < 0.001): 89.3 ± 28.8 nmol/L vs. 145.2 ± 42.7 nmol/L, respectively.
Heald	Blood samples on LT4: No information.Blood samples on DTE treatment: 2–4 hours after administration of DTE.	TSH^c values before and during DTE are presented but not compared: 2.1 ± 0.85 (range: 0.2–9.8) vs. 0.8 ± 1.8 (range: <0.01–8.5) mIU/L, respectively.	No fT3 levels when patients were treated with LT4 are available.fT3^c during DTE: 7.9 ± 3.4 pmol/L (range: 4.0–11.9).	fT4^c values before and during DTE are presented but not compared: 18.9 ± 4.2 (range: 14.6–23.3) vs. 13.2 ± 4.4 pmol/L (range: 8.3–30), respectively.
Hoang	In the morning, fasting, and before administration of thyroid medication.	TSH^d was higher on DTE than on LT4 (p = 0.0032): 1.67 ± 0.77 vs. 1.30 ± 0.63 mIU/L, respectively.	TT3^d was higher on DTE than on LT4 ( p < 0.0001): 138.96 ± 47.26 ng/dL vs. 89.13 ± 19.48 ng/dL, respectively.	TT4^d was lower on DTE than on LT4 ( p < 0.0001): 5.88 ± 1.34 µg/dL vs. 9.26 ± 2.05 µg/dL, respectively.fT4^d was lower on DTE than on LT4 ( p < 0.0001): 0.85 ± 0.16 ng/dL vs. 1.36 ± 0.27 ng/dL, respectively.
Kuye	No information.	TSH^e was lower on DTE than on LT4 ( p = 0.001): 2.4 (0.18) and 3.4 (0.27) mIU/mL, respectively. There was no difference between treatments in percent of TSH within range, or in “percent of patients with all TSH within range.” Visit-to-visit TSH variability was lower on LT4 than on DTE ( p = 0.015).
Lee	No information.	TSH^f did not change over time and were not different between DTE and LT4 regimens (p value not given).		fT4 index^f decreased over time on DTE ( p < 0.01). Serum TT4 levels^f were lower on DTE ( p < 0.05).
Pepper	In the morning after administration of thyroid medication.	There was no difference in TSH^g between DTE and LT4 regimens: 1.34 ± 2.4 vs. 1.30 ± 1.7 mIU/L, respectively (p value not given).	Serum T3^g was higher on DTE than on LT4 ( p < 0.003): 1.50 ± 0.86 vs. 1.18 ± 0.51 pmol/L.	Serum T4^g levels during the LT4 and DTE regimens are presented in the article but not compared.
Pepper	In the morning after administration of thyroid medication.		T4:T3 ratio was lower on DTE than on LT4 ( p < 0.001):4.70 ± 2.0 vs. 8.45 ± 3.7, respectively.
Shakir	In the morning, fasting, and before administration of thyroid medication.	TSH^h was higher on DTE than on LT4 ( p < 0.001) but remained within normal range: 2.34 (1.48) vs. 1.63 (1.28) mIU/L, respectively.	TT3^h was higher on DTE than on LT4 (p < 0.001): LT4: 103 (22.1) ng/dL vs. DTE: 155 (48.7) ng/dL.	TT4^h was lower on DTE than on LT4 ( p < 0.001): 5.62 (1.38) vs. 8.46 (2.14) µg/dL, respectively.fT4^h was lower on DTE than on LT4 ( p < 0.001): 0.980 (0.673) vs. 1.44 (0.358) ng/dL.
			TT4:TT3 ratio was lower during DTE compared with LT4 ( p < 0.001): 36.2 (2.8) and 82.1 (9.7), respectively.
Surks	Blood samples after DTE intake.		After DTE administrationⁱ: Plasma T3 was unchanged in one patient and increased 100% in two patients.	After DTE administrationⁱ: Plasma T4 was relatively constant throughout the 24 hours.
Tariq	No information.	TSH^j remained in the reference range more often on DTE than on LT4 ( p = 0.0082): 82.76% TSH in range during LT4 and 96.49% during DTE.	fT3^j remained in the reference range more often on DTE than on LT4 ( p = 0.0126): 74.47% fT3 in range during LT4 and 93.62% during DTE.	No difference between DTE and LT4 in percent of fT4^j within range (p = 1.00).

No statistical methods used to compare groups.

TSH reference range: <4.0 mIU/L (lower detection limit was 0.6 mIU/L, i.e., the method could not distinguish between euthyroidism and thyrotoxicosis). T3 reference range: 1.5–3.1 nmol/L. T4 reference range: 80–180 nmol/L.

TSH reference range: 0.35–5.50 mIU/L. fT3 reference range: 3.5–6.5 pmol/L. fT4 reference range: 8.3–30.0 pmol/L.

TSH reference range: 0.27–4.20 mIU/L. TT3 reference range: 60–181 ng/dL. TT4 reference range: 4.5–12 µg/dL. fT4 reference range: 0.89–1.76 ng/dL.

TSH reference range: 0.32–5.5 mIU/L.

No reference range is specified in the article.

TSH reference range: 0.4–4.0 mIU/L. Reference range was not specified for T4 and T3.

p-value reports the results of a linear mixed-effects model evaluation of the fixed effect of the three treatment regimes, using participants as a random effect. TSH reference range: 0.27–4.20 mIU/L. TT3 reference range: 60–181 ng/dL. TT4 reference range: 4.5–12 µg/dL. fT4 reference range: 0.89–1.76 ng/dL. T4:T3 ratio reference range: 22.7–150.

DTE was administered to three patients: two received 120 mg DTE, and one received 60 mg DTE. Serum was obtained at intervals of 1–2 hours throughout the day. No p-values are given.

Mean TSH, fT3, and fT4 values are presented during DTE and LT4 therapy but without comparing regimens. TSH reference range: 0.35–5.5 mIU/L. Patients with an abnormally low TSH had previous thyroid cancer. Thyroid status was measured in 60 patients receiving LT4 and 57 receiving DTE.

DTE, desiccated thyroid extract; fT3, free triiodothyronine; fT4, free thyroxine; LT4, levothyroxine; TT4, total thyroxine; TT3, total triiodothyronine.

In the study by Baisier et al., 24-hour urine excretion of T3 was higher during DTE treatment compared with LT4 or no treatment.³⁶ Likewise, several studies found higher serum/plasma T3 levels on DTE treatment compared with LT4.^32
–34,45 In the study by Tariq et al., DTE was better than LT4 in maintaining fT3 within the reference range, in line with the finding for TSH.³⁷ Finally, Surks et al. investigated, in three patients, changes in plasma T3 after ingestion of DTE; in two participants, plasma T3 increased by 100%, while one patient showed no change.³⁸

Plasma T4 (measured as free T4, total T4, or free T4 index) was lower on DTE than on LT4 treatment in all studies evaluating thyroid function measures (Table 4). Surks et al. demonstrated that plasma T4 was stable throughout 24 hours on DTE treatment.³⁸ Furthermore, in the study by Tariq et al., plasma T4 was within target to the same extent when comparing DTE and LT4 treatment.³⁷

Lee et al. found small fluctuations in plasma levels of thyroid peroxidase and thyroglobulin antibodies when treated with DTE or LT4, but without any difference between the two treatment regimens.⁴³ No other studies included data on thyroid autoantibodies.

Cardiovascular markers and anthropometric measures

The effect on lipids was evaluated only in the two RCTs, with slightly different results.^32,33 In the two-armed RCT, high-density lipoprotein (HDL) was significantly lower in DTE-treated patients compared with those receiving LT4 (Table 5), while Shakir et al. found no difference.^32,33 In both RCTs, low-density lipoprotein (LDL), total cholesterol, and triglyceride levels were unaffected by the type of treatment (Table 5).^32,33 In a fixed-effects meta-analysis comparing DTE with LT4, the overall Hedges’ g of triglyceride was −0.02 (95% CI: −0.25, 0.21; p = 0.84), for LDL −0.08 (95% CI: −0.31, 0.15; p = 0.48), for total cholesterol −0.05 (95% CI: −0.28, 0.18; p = 0.65), and for HDL −0.05 (95% CI: −0.28, 0.18; p = 0.69) (Supplementary Data S6). The heterogeneity was low in all outcomes (I ² = 0%) (Supplementary Data S6).

Table 5.

Association Between Treating Hypothyroidism with Desiccated Thyroid Extract and Body Mass Index, Body Weight, Lipid Profile, Heart Rate, and Blood Pressure

1^st author	BMI/body weight (mean ± standard deviation)	Lipid profile (mean ± standard deviation)	Resting heart rate (mean ± standard deviation)	Blood pressure
Heald	No difference in mean BMI before and after DTE treatment^a	—	—	—
Hoang	Decrease of 1.30 kg^b with DTE treatment compared with LT4 ( p < 0.0001)- Baseline: 79.1 ± 16.6 kg^b - After LT4: 79.7 ± 17.1 kg^b - After DTE: 78.4 ± 16.5 kg^b	HDL was significantly lower after DTE treatment compared with LT4 ( p = 0.028)- After LT4: 63.22 ± 15.25^c mg/dL- After DTE: 60.97 ± 15.16^c mg/dLTotal cholesterol (p = 0.105), LDL (p = 0.418), and triglyceride (p = 0.176) were similar after LT4 and DTE treatment.	No difference in heart rate between DTE and LT4 (p = 0.587).	No difference between DTE and LT4 (systolic: p = 0.342, diastolic: p = 0.475).
Pepper	No difference in weight before and after DTE treatment (p value not available).	—	—	—
Shakir	No difference in weight during treatment with DTE, LT4, and LT4/LT3 (p = 0.278).	No difference in HDL (p = 0.773), LDL (p = 0.31), total cholesterol (p = 0.85), or triglyceride (p = 0.411) after treatment with DTE, LT4, and LT4/LT3.	Increased heart rate when treated with DTE compared with LT4 and LT4/LT3 ( p = 0.04):	No difference between DTE, LT4 and LT4/LT3 (systolic: p = 0.13, diastolic: p = 0.64).
			- DTE: 73.7 ± 14.3 beats/min
			- LT4: 71.5 ± 11.1 beats/min
			- LT4/LT3: 70.4 ± 14.6 beats/min

In 17 participants, BMI was measured when initiating DTE and at their last follow-up visit.

Converted from pounds to kilograms (kg).

Normal range > 40 mg/dL.

BMI, body mass index; DTE, desiccated thyroid extract; HDL, high-density lipoprotein; LDL, low-density lipoprotein; LT4, levothyroxine; LT3, liothyronine; min, minute.

Regarding body weight, neither Heald et al. nor Pepper et al. found any difference in mean body mass index before and after treatment with DTE (Table 5).^35,45 The two RCTs showed divergent results, as Hoang et al. found a significant decrease of 1.3 kg (p < 0.0001) in body weight during treatment with DTE, compared with LT4, while Shakir et al. found neutral effects (Table 5).^32,33 In a fixed-effects meta-analysis, there was no difference in body weight when comparing DTE with LT4 (p = 0.67), and the heterogeneity was low (I ² = 0%) (Supplementary Data S6).

In the RCT by Shakir et al., patients had an increased heart rate when treated with DTE, compared with LT4 and LT4/LT3 (p = 0.04).³³ In contrast, the heart rate was unaffected by type of treatment in the other RCT by Hoang et al. (Table 5, Supplementary Data S6).³² None of the RCTs found any association between blood pressure and treatment regimen, and the meta-analysis showed no difference comparing DTE with LT4 treatment (Table 5, Supplementary Data S6).^32,33

Adverse events

Adverse effect was an outcome in the study by Tariq et al.³⁷ One patient discontinued DTE due to mild hyperthyroid symptoms. No participants were hospitalized for adverse events or arrhythmias, but 6.7% of the patients (whether treated with DTE or LT4/LT3) complained of palpitations. All these individuals had a plasma TSH level <0.35 μIU/L. In the study by Penna et al.,³⁹ fewer patients treated with DTE (2.8%) or LT4/LT3 (5.8%), compared with LT4 (7.7%), had outpatient visits associated with cardiovascular or bone disease (p = 0.023), but the difference was insignificant when adjusted for sociodemographic and clinical covariates. There was no difference in emergency or hospital visits.³⁹

Five other articles reported adverse effects but not as a prespecified outcome.^{32,33,35,43,45} In the study by Heald et al., two participants discontinued DTE due to palpitations but without abnormalities on ECG.³⁵ In the study by Pepper et al., three patients on DTE had complaints of itchiness of the eyelids, scalp hair loss, and palpitations, respectively.⁴⁵ Lee et al., Shakir et al., and Hoang et al. did not observe any adverse events among their participants.^32,33,43

The three included case reports also described adverse effects due to over-the-counter bovine thyroid supplementations or high-dose DTE treatment, respectively (Table S1).^40
–42 These adverse effects were all due to iatrogenic thyrotoxicosis and included atrial fibrillation, weight loss, tachycardia, and myocardial infarction (Table S1).^40
–42

Treatment preference

In the RCT by Hoang et al., 34 patients (48.6%) preferred DTE, 13 (18.6%) preferred LT4, and 23 (32.9%) had no preference.³² Preference for DTE over LT4 was statistically significant (p = 0.002). In the RCT by Shakir et al., no difference was found in preference among the three treatment regimens, that is, DTE, LT4, and LT4/LT3.³³

In the unblinded study by Pepper et al., a five-point satisfaction rating scale (SRS) was used to evaluate treatment preference between LT4 and DTE. The scale ranged from 5 points (“DTE definitely superior to LT4”) to 1 point (“DTE definitely worse than LT4”).⁴⁵ Of the 154 DTE-treated patients, 117 patients (78%) indicated an SRS score >3 (responders), while 33 patients (22%) indicated a score ≤3 (nonresponders) (p < 0.001).⁴⁵

Gene polymorphism

The participants in the study by Shakir et al. were examined for DIO2 polymorphism, that is, the Ala92-DIO2 and Thr92-DIO2 variants.³³ Of 75 participants, 41 were heterozygotes, while 9 and 25 individuals were monozygotic for the Ala92-DIO2 and Thr92-DIO2 variants, respectively. The authors found no difference in the evaluated outcomes, that is, GHQ-12, TSQ-36, BDI, and WMS-IV, when the gene variants were considered.³³

Discussion

In this systematic review, there was a dearth of high-quality studies. We chose to include all types of studies, realizing that the published reports on DTE treatment in hypothyroid adults were limited. Despite this broad search strategy, only 11 studies and 3 case reports were included, here among two blinded RCTs. These latter two trials were generally well-conducted, only judged with some concerns due to the lack of an a priori statistical plan.

Different doses of DTE, nonstandardized treatment regimens, various DTE preparations and brands, and patient selection are all factors that complicate comparison across studies. It is reported that a DTE dose of 80 mg on average should restore clinical and biochemical euthyroidism.⁴⁷ However, the participants received DTE doses in the range of 200–233 mg in one NRSI, while only 30 mg DTE per day was given in another study.^36,37 These huge differences may reflect that some patients were over- or undertreated, respectively, or that the participants’ thyroid functional reserve differed between studies. In fact, information on the proportion of subclinical and overt hypothyroid patients is not provided in virtually any of the studies.

Quality-of-life and symptom scores

In the two RCTs, no difference in QoL or symptom scores was found by comparing a DTE regimen with either LT4 or LT4/LT3 combination therapy.^32,33 On the contrary, in the NRSIs by Heald et al. and Tariq et al., QoL as well as symptom scores improved significantly when patients were switched from LT4 monotherapy to DTE because of persistent symptoms.^35,37 The discrepancy in outcome between the RCTs and the NRSIs is most likely caused by the nonblinded design of the latter studies. Indeed, patients with hypothyroidism are highly susceptible to the placebo effect, as demonstrated in the recent RCT of the effect of selenium supplementation.⁴⁹

Various tools for assessment of QoL were used, which makes comparison across studies difficult. The GHQ-12, used in both RCTs, has been extensively validated in different settings, but not in thyroid patients, to the best of our knowledge.^46,50 The same reservation goes for SF-20 used by Tariq et al.^37,48 The ThyPRO, used by Heald et al., has been developed and validated specifically for patients with benign thyroid disease, and this tool seems the obvious choice for estimating the QoL in future studies of this patient group.^35,51,52

Thyroid hormone levels

Since T3 has a short half-life, multiple daily dosing is required to restore a near-normal diurnal serum profile of T3. In most studies included in the present review, the patients were given DTE only once daily. Nevertheless, serum T3 in patients treated with DTE largely remained within the normal range, even though at a higher level when compared with LT4 monotherapy. The levels of TSH also remained within the normal range but were more fluctuating during DTE therapy. Kuye et al. showed no difference between DTE and LT4 treatment in TSH values within range over a three-year period.⁴⁴ In contrast to other studies, Kuye et al. adjusted for a range of potential confounders. However, for unknown reasons, patients with Hashimoto’s thyroiditis were excluded in that study.⁴⁴

Adverse effects

Long-term studies of patients treated with DTE are lacking. Regarding short-term adverse effects, DTE—compared with treatment with LT4—seems to carry a higher risk of cardiac events and symptoms of thyrotoxicosis. However, the overall level of evidence was judged as very low in line with other outcomes of this review. It is most likely that the majority of adverse effects seen with DTE therapy are caused by iatrogenic thyrotoxicosis, which also is the perception among physicians and patients.^53,54 However, transient iatrogenic and supraphysiological levels of T3 during DTE treatment may raise concern. Guidelines advocate for the normalization of TSH serum levels, as TSH outside the normal range is associated with increased risk of cardiac events and mortality.³ A similar recommendation does not exist for T3, even though increased levels of T3 have been linked to cancer development and cardiovascular events.^3,6,8 Importantly, as serum T3 shows diurnal variations in patients on LT3 treatment, T3 levels that periodically exceed the upper reference limit may remain undetected by the conventional biochemical monitoring.⁵⁵ Therefore, blood sampling for measurement of serum T3 should be standardized to be clinically useful. The same applies to future long-term studies of patients treated with DTE.

Treatment preference

According to an online survey, patients treated with DTE claimed higher treatment satisfaction compared with other treatment regimens.⁵⁶ Such a result is without a doubt under the major influence from selection bias. Significantly more patients preferred DTE to LT4 in the RCT by Hoang et al.,³² but the preference for DTE was not significant compared with patients who indicated “no preference.” In the RCT by Shakir et al., no preference was found overall, but when dividing the participants according to severity of symptoms on LT4 treatment, the most symptomatic participants seemed to prefer DTE or LT4/LT3.^33,57 Several theories have been proposed as to why some patients prefer DTE to LT4. Involved factors may include enhanced well-being from transient thyrotoxicosis, or postulated positive effects of other ingredients in DTE, for example, 3,5-diiodothyronine and calcitonin.⁵⁸

DIO2 gene polymorphisms

The presence of the Thr92Ala gene variant in the DIO2 gene may result in reduced conversion of T4 into T3 in athyreotic patients on LT4, potentially leading to reduced levels of T3 in certain tissues, such as the brain.⁵⁹ Indeed, previous studies have indicated that individuals with the Thr92Ala gene variant are those who benefit most from the LT4/LT3 combination therapy.^60,61 Only one study investigated whether improvement in QoL depended on the presence of this gene variant, but no such association was demonstrated.³³

Strengths and limitations

We used a systematic approach and did a comprehensive literature search in several databases. The quality assessment of the included studies was done with appropriate tools. We judged some older studies outdated and unsuitable for the present review due to unreliable methods for diagnostics and monitoring. In addition, before 1985, the amount of thyroid hormone in DTE was estimated by measuring the iodine content, with the result that tablets across batches had various potency, and some were even devoid of any active ingredient.^47,62 Nevertheless, some of the studies we decided to exclude have been well cited in the past and made up the cornerstones in previous evidence supporting the use of DTE.

Future perspectives

There is an ongoing debate about the mechanisms behind the persistent symptoms in some patients on LT4, and how these individuals should be treated and managed.^12,56 In future studies, preferably in the setting of RCTs, the focus should be on patients with persistent symptoms, low circulating T3 levels, or those with specific DIO2 gene variants. Likewise, the assessment tools for estimating QoL should be well-validated and disease-specific.

Considering the short half-life of T3, future studies should aim to administer DTE at least twice daily to achieve more physiological T3 levels. Due to the potentially negative impact of the high T3 content, it is of crucial importance that future studies of DTE treatment include long-term follow-up of patients to disclose any slow-onset adverse effects. At least, surrogate biomarkers that reflect unintended effects on other organs should be measured.

Until more solid evidence is provided, the management of patients with hypothyroidism with persistent symptoms remains challenging, as some people may rely more on web-based information than on their caretaking physician.^54,56 This increases the risk of overtreatment due to self-adjusted treatment, and over-the-counter thyroid preparations.^16,40 To address this problem, education and information of patients should be encouraged, and the development and validation of different self-management tools and point-of-care devices are also likely to prove very helpful.

Conclusions

This review of DTE treatment in patients with hypothyroidism comprised 14 previous studies, the majority of which suffered from a poor study design. Accordingly, the quality of current evidence on effects and adverse effects of DTE treatment is low. Several studies indicated that patients with persistent symptoms show improvements when treated with DTE, compared with LT4 monotherapy, but the conclusions are hampered by the lack of a control group, and the positive results most likely reflect a placebo effect. Two well-designed blinded RCTs, comparing DTE with LT4 and/or LT4/LT3 combination therapy, were unable to demonstrate any convincing differences in QoL or symptom scores between the treatment regimens.

Footnotes

Acknowledgments

Research librarian Mette B. Eriksen at the University Library of Southern Denmark is acknowledged for her help in developing the search strategy. Librarians at the University Library of Southern Denmark are acknowledged for providing the full text of several screened articles. The Centre for Evidence Based Medicine Odense (CEBMO) is acknowledged for helping with protocol writing and providing guidance on risk-of-bias assessments.

Authors’ Contributions

K.R.R. contributed to the conceptualization, data curation, formal analysis, funding acquisition, investigation (screening of references and risk-of-bias assessment), project administration, software, validation, visualization, and writing (original draft, review & editing). C.B.L. contributed to the investigation (screening of references and risk-of-bias assessment) and writing (review & editing). S.T.B. contributed to conceptualization, funding acquisition, project administration, resources, supervision, validation, and writing (review & editing).

Author Disclosure Statement

K.R.R., C.B.L., and S.J.B. declare no conflicts of interest.

Funding Information

K.R.R. is enrolled as a Ph.D. student at the University of Southern Denmark and has received research grants from the Danish Medicines Agency, the Region of Southern Denmark (Grant number: 20/14658), and a grant from the “Board of Chief Physicians at Odense University Hospital” (Grant number: A4641). C.B.L. is enrolled as a Ph.D. student at the University of Southern Denmark and has received institutional research grants from the University of Southern Denmark and the Region of Southern Denmark.

Supplementary Material

Supplementary Data S1

Supplementary Data S2

Supplementary Data S3

Supplementary Data S4

Supplementary Data S5

Supplementary Data S6

References

McAninch

, Bianco

. The swinging pendulum in treatment for hypothyroidism: From (and toward?) combination therapy. Front Endocrinol (Lausanne), 2019; 10:446; doi: 10.3389/fendo.2019.00446

Hennessey

, Espaillat

. Current evidence for the treatment of hypothyroidism with levothyroxine/levotriiodothyronine combination therapy versus levothyroxine monotherapy. Int J Clin Pract, 2018; 72(2):1–14; doi: 10.1111/ijcp.13062

Jonklaas

, Bianco

, Bauer

, et al. American Thyroid Association Task Force on Thyroid Hormone Replacement. Guidelines for the treatment of hypothyroidism: Prepared by the American thyroid association task force on thyroid hormone replacement. Thyroid, 2014; 24(12):1670–1751; doi: 10.1089/thy.2014.0028

Abrahamsen

, Jørgensen

, Laulund

, et al. The excess risk of major osteoporotic fractures in hypothyroidism is driven by cumulative hyperthyroid as opposed to hypothyroid time: An observational register-based time-resolved cohort analysis. J Bone Miner Res, 2015; 30(5):898–905; doi: 10.1002/jbmr.2416

Bano

, Chaker

, Mattace-Raso

FUS

, et al. Thyroid function and the risk of atherosclerotic cardiovascular morbidity and mortality: The Rotterdam study. Circ Res, 2017; 121(12):1392–1400; doi: 10.1161/CIRCRESAHA.117.311603

Lillevang-Johansen

, Abrahamsen

, Jørgensen

, et al. Duration of over- and under-treatment of hypothyroidism is associated with increased cardiovascular risk. Eur J Endocrinol, 2019; 180(6):407–416; doi: 10.1530/EJE-19-0006

, Kim

, et al. Heart failure and stroke risks in users of liothyronine with or without levothyroxine compared with levothyroxine alone: A propensity score-matched analysis. Thyroid, 2022; 32(7):764–771; doi: 10.1089/thy.2021.0634

Tosovic

, Bondeson

, et al. T3 levels in relation to prognostic factors in breast cancer: A population-based prospective cohort study. BMC Cancer, 2014; 14(536):536; doi: 10.1186/1471-2407-14-536

Planck

, Hedberg

, Calissendorff

, et al. Liothyronine use in hypothyroidism and its effects on cancer and mortality. Thyroid, 2021; 31(5):732–739; doi: 10.1089/thy.2020.0388

10.

Leese

, Soto-Pedre

, Donnelly

. Liothyronine use in a 17 year observational population-based study-the tears study. Clin Endocrinol (Oxf), 2016; 85(6):918–925; doi: 10.1111/cen.13052

11.

Cappola

. Design of the optimal trial of combination therapy. Front Endocrinol (Lausanne), 2020; 11(April):168; doi: 10.3389/fendo.2020.00168

12.

Saravanan

, Chau

, Roberts

, et al. Psychological well-being in patients on “adequate” doses of L-thyroxine: Results of a large, controlled community-based questionnaire study. Clin Endocrinol (Oxf), 2002; 57(5):577–585; doi: 10.1046/j.1365-2265.2002.01654.x

13.

Djurovic

, Pereira

, Smit

JWA

, et al. Cognitive functioning and quality of life in patients with Hashimoto thyroiditis on long-term levothyroxine replacement. Endocrine, 2018; 62(1):136–143; doi: 10.1007/s12020-018-1649-6

14.

Winther

, Cramon

, Watt

, et al. Disease-specific as well as generic quality of life is widely impacted in autoimmune hypothyroidism and improves during the first six months of levothyroxine therapy. PLoS One, 2016; 11(6):e0156925; doi: 10.1371/journal.pone.0156925

15.

Peterson

, McAninch

, Bianco

. Is a normal TSH synonymous with “euthyroidism” in levothyroxine monotherapy? J Clin Endocrinol Metab, 2016; 101(12):4964–4973; doi: 10.1210/jc.2016-2660

16.

Michaelsson

, Medici

, la Cour

, et al. Treating hypothyroidism with thyroxine/triiodothyronine combination therapy in Denmark: Following guidelines or following trends? Eur Thyroid J, 2015; 4(3):174–180; doi: 10.1159/000437262

17.

Riis

, Frølich

, Hegedüs

, et al. Use of thyroid hormones in hypothyroid and euthyroid patients: A 2020 Thesis questionnaire survey of members of the Danish Endocrine Society. J Endocrinol Invest, 2021; 44(11):2435–2444; doi: 10.1007/s40618-021-01555-y

18.

Buffet

, Belin

, Attanasio

, et al. Real-life practice of thyroid hormone use in hypothyroid and euthyroid patients: A detailed view from the Thesis questionnaire survey in France. Ann Endocrinol (Paris), 2022; 83(1):27–34; doi: 10.1016/j.ando.2021.11.002

19.

Vardarli

, Brandenburg

, Hegedüs

, et al. A questionnaire survey of German thyroidologists on the use of thyroid hormones in hypothyroid and euthyroid patients: The Thesis collaborative. Exp Clin Endocrinol Diabetes, 2022; 130(9):577–586; doi: 10.1055/a-1832-0644

20.

Borissova

, Boyanov

, Attanasio

, et al. Use of thyroid hormones in hypothyroid and euthyroid patients: A THESIS* questionnaire survey of Bulgarian physicians. *THESIS: Treatment of Hypothyroidism in Europe by Specialists: An International Survey. Endocrinologia, 2020; 25(4):299–309.

21.

Burlacu

, Attanasio

, Hegedüs

, et al. Use of thyroid hormones in hypothyroid and euthyroid patients: A thesis* survey of Belgian specialists. Thyroid Res, 2022; 15(1):3–309; doi: 10.1186/s13044-022-00121-9

22.

Jiskra

, Paleček

, Attanasio

, et al. Use of thyroid hormones in hypothyroid and euthyroid patients: A 2020 thesis questionnaire survey of members of the Czech Society of Endocrinology. BMC Endocr Disord, 2022; 22(1):117; doi: 10.1186/s12902-022-01027-1

23.

Planck

, Lantz

, Perros

, et al. Use of thyroid hormones in hypothyroid and euthyroid patients: A 2020 THESIS questionnaire survey of members of the Swedish Endocrine Society. Front Endocrinol (Lausanne), 2021; 12:795111; doi: 10.3389/fendo.2021.795111

24.

Mustafa

, Ali

, McGowan

, et al. Use of thyroid hormones in hypothyroid and euthyroid patients: A thesis questionnaire survey of members of the Irish Endocrine Society. Ir J Med Sci, 2022; 192(5):2179–2187; doi: 10.1007/s11845-022-03235-z

25.

Moher

, Shamseer

, Clarke

, et al. PRISMA-P Group. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev, 2015; 4(1):1–9; doi: 10.1186/2046-4053-4-1

26.

Riis

, Larsen

, Gregersen

, et al. Potential benefits and risks by using desiccated thyroid extract for treatment of hypothyroidism: Protocol for a systematic review. Prospero, 2020:1–7.

27.

McAninch

, Bianco

. The history and future of treatment of hypothyroidism. Ann Intern Med, 2016; 164(1):50–56; doi: 10.7326/M15-1799

28.

Hennessey

. Historical and current perspective in the use of thyroid extracts for the treatment of hypothyroidism. Endocr Pract, 2015; 21(10):1161–1170; doi: 10.4158/EP14477.RA

29.

Sterne

, Hernán

, Reeves

, et al. ROBINS-I: A tool for assessing risk of bias in non-randomised studies of interventions. Bmj, 2016; 355:i4919; doi: 10.1136/bmj.i4919

30.

Higgins

, Savović

, Page

, et al. Revised Cochrane risk-of-bias tool for randomized trials (RoB 2). Cochrane, 2019(August):1–72.

31.

Balshem

, Helfand

, Schünemann

, et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol, 2011; 64(4):401–406; doi: 10.1016/j.jclinepi.2010.07.015

32.

Hoang

, Olsen

, Mai

, et al. Desiccated thyroid extract compared with levothyroxine in the treatment of hypothyroidism: A randomized, double-blind, crossover study. J Clin Endocrinol Metab, 2013; 98(5):1982–1990; doi: 10.1210/jc.2012-4107

33.

Shakir

MKM

, Brooks

, Mcaninch

, et al. Comparative effectiveness of levothyroxine, desiccated thyroid extract, and levothyroxine+liothyronine in hypothyroidism. J Clin Endocrinol Metab, 2021; 106(11):E4400–E4413; doi: 10.1210/clinem/dgab478

34.

Bindels

AJCH

, Meinders

. The serum concentrations of T3, T4 and TSH in evaluating replacement therapy in primary hypothyroidism. Neth J Med, 1988; 32(1-2):59–71.

35.

Heald

, Premawardhana

, Taylor

, et al. Is there a role for natural desiccated thyroid in the treatment of levothyroxine unresponsive hypothyroidism? Results from a consecutive case series. Int J Clin Pract, 2021; 75(12):e14967; doi: 10.1111/ijcp.14967

36.

Baisier

, Hertoghe

, Eeckhaut

. Thyroid insufficiency. Is thyroxine the only valuable drug? J Nutr Environ Med, 2001; 11(3):159–166; doi: 10.1080/13590840120083376

37.

Tariq

, Wert

, Cheriyath

, et al. Effects of long-term combination LT4 and LT3 therapy for improving hypothyroidism and overall quality of life. South Med J, 2018; 111(6):363–369; doi: 10.14423/SMJ.0000000000000823

38.

Surks

, Schadlow

, Oppenheimer

. A new radioimmunoassay for plasma L-triiodothyronine: Measurements in thyroid disease and in patients maintained on hormonal replacement. J Clin Invest, 1972; 51(12):3104–3113; doi: 10.1172/JCI107137

39.

Penna

, Bianco

, Ettleson

. A cross-sectional analysis of cardiovascular and bone healthcare utilization during treatment with thyroid hormone. J Clin Endocrinol Metab, 2023; 109(3):e1143–e1150; doi: 10.1210/clinem/dgad629

40.

Mohebbi

, Chen

. A ‘Natural’ thyroid storm!. J Clin Pharm Ther, 2019; 44(5):813–814; doi: 10.1111/jcpt.12996

41.

Sawin

, London

. “Natural” desiccated thyroid. A “health-food” thyroid preparation. Arch Intern Med, 1989; 149(9):2117–2118; doi: 10.1001/archinte.149.9.2117

42.

Bergeron

, Goldsmith

, Schiller

. Myocardial infarction, severe reversible ischemia, and shock following excess thyroid administration in a woman with normal coronary arteries. Arch Intern Med, 1988; 148(6):1450–1453; doi: 10.1001/archinte.1989.00390030147029

43.

Lee

, Scherberg

, DeGroot

. Induction of oral tolerance in human autoimmune thyroid disease. Thyroid, 1998; 8(3):229–234; doi: 10.1089/thy.1998.8.229

44.

Kuye

, Riggs

, King

, et al. Thyroid stimulating hormone stability in patients prescribed synthetic or desiccated thyroid products: A retrospective study. Ann Fam Med, 2020; 18(5):452–454; doi: 10.1370/afm.2545

45.

Pepper

, Casanova-Romero

. Conversion to Armour thyroid from levothyroxine improved patient satisfaction in the treatment of hypothyroidism. J Endocrinol Diabetes Obes, 2014; 2(3):2–6.

46.

Pevalin

. Multiple applications of the GHQ-12 in a general population sample: An investigation of long-term retest effects. Soc Psychiatry Psychiatr Epidemiol, 2000; 35(11):508–512; doi: 10.1007/s001270050272

47.

Idrees

, Palmer

, MacIel

RMB

, et al. Liothyronine and desiccated thyroid extract in the treatment of hypothyroidism. Thyroid, 2020; 30(10):1399–1413; doi: 10.1089/thy.2020.0153

48.

Stewart

, Hays

Jr , Ware

. The MOS short-form general health survey: Reliability and validity in a patient population. Med Care, 1988; 26(7):724–735.

49.

Larsen

, Winther

, Cramon

, et al. Selenium supplementation and placebo are equally effective in improving quality of life in patients with hypothyroidism. Eur Thyroid J, 2024; 13(1); doi: 10.1530/ETJ-23-0175

50.

Cramon

, Bjorner

, Groenvold

, et al. Implementation of thyroid-related patient-reported outcomes in routine clinical practice. Front Endocrinol (Lausanne), 2022; 13(September):1000682–1000610; doi: 10.3389/fendo.2022.1000682

51.

Watt

, Cramon

, Hegedüs

, et al. The thyroid-related quality of life measure ThyPRO has good responsiveness and ability to detect relevant treatment effects. J Clin Endocrinol Metab, 2014; 99(10):3708–3717; doi: 10.1210/jc.2014-1322

52.

Watt

, Hegedüs

, Groenvold

, et al. Validity and reliability of the novel thyroid-specific quality of life questionnaire, ThyPRO. Eur J Endocrinol, 2010; 162(1):161–167; doi: 10.1530/EJE-09-0521

53.

Shrestha

, Malabanan

, Haugen

, et al. Adverse event reporting in patients treated with thyroid hormone extract. Endocr Pract, 2017; 23(5):566–575; doi: 10.4158/EP161584.OR

54.

Toloza

FJK

, Suarez

NRE

, El Kawkgi

, et al. Patient experiences and perceptions associated with the use of desiccated thyroid extract. Med, 2020; 56(4):1–12.

55.

Saravanan

, Siddique

, Simmons

, et al. Twenty-four hour hormone profiles of TSH, free T3 and free T4 in hypothyroid patients on combined T3/T4 therapy. Exp Clin Endocrinol Diabetes, 2007; 115(4):261–267; doi: 10.1055/s-2007-973071

56.

Peterson

, Cappola

, Castro

, et al. An online survey of hypothyroid patients demonstrates prominent dissatisfaction. Thyroid, 2018; 28(6):707–721; doi: 10.1089/thy.2017.0681

57.

Bonnema

, Riis

, Jensen

. Bonnema et al: “comparative effectiveness of levothyroxine, desiccated thyroid extract, and Levothyroxine+Liothyronine in Hypothyroidism. J Clin Endocrinol Metab, 2022; 107(3):E1317–E1318; doi: 10.1210/clinem/dgab778

58.

Perros

, Hegedus

. Enhanced well-being associated with thyrotoxicosis: A neglected effect of thyroid hormones? Int J Endocrinol Metab, 2022; 20(2):e127230; doi: 10.5812/ijem-127230

59.

Castagna

, Dentice

, Cantara

, et al. DIO2 Thr92Ala reduces Deiodinase-2 activity and serum-T3 levels in thyroid-deficient patients. J Clin Endocrinol Metab, 2017; 102(5):1623–1630; doi: 10.1210/jc.2016-2587

60.

Carlé

, Faber

, Steffensen

, et al. Hypothyroid patients encoding combined MCT10 and DIO2 gene polymorphisms may prefer L-T3 + L-T4 combination treatment–data using a blind, randomized, clinical study. Eur Thyroid J, 2017; 6(3):143–151; doi: 10.1159/000469709

61.

Panicker

, Saravanan

, Vaidya

, et al. Common variation in the DIO2 gene predicts baseline psychological well-being and response to combination thyroxine plus triiodothyronine therapy in hypothyroid patients. J Clin Endocrinol Metab, 2009; 94(5):1623–1629; doi: 10.1210/jc.2008-1301

62.

Mateo

RCI

, Hennessey

. Thyroxine and treatment of hypothyroidism: Seven decades of experience. Endocrine, 2019; 66(1):10–17; doi: 10.1007/s12020-019-02006-8