Keywords
subclinical hypothyroidism, levothyroxine, pregnancy, adults, meta-analysis, evidence synthesis
The prevalence of subclinical hypothyroidism (SCH) in the general population is notably high, ranging from 4% to 20%, and varies according to sex and age. SCH has been reported to be associated with multiple adverse pregnancy outcomes in pregnant women. While the therapeutic decision for overt hypothyroidism is fairly straightforward, treatment of SCH, especially the timing of the initiation of therapy, has been a point of discussion.
To assess the effect of early initiation of treatment compared to watchful waiting, delayed initiation, or no treatment on progression to overt hypothyroidism, prevention of complications, and associated adverse outcomes through a "systematic review and meta-analysis (SRMA)”.
The SRMA protocol adhered to the “Preferred Reporting Items for Systematic reviews and Meta-Analyses Protocols (PRISMA-P 2015)” guidelines. Only “randomized controlled trials (RCTs)” will be included. The databases “PubMed, Scopus, EMBASE, and the Cochrane Library” will be searched from inception until 30.12.2025. Two stages (title abstract followed by full-text review) and two-pass screening (two authors independently) with a third reviewer adjudication of conflicts will be adopted. Data will be extracted, and the risk of bias will be assessed independently by two authors using a process to resolve differences. The risk of bias for RCTs will be assessed using Cochrane ROB 2.0. Pooled estimates will be calculated for meta-analysis. Subgroup analysis and meta-regression will be performed if heterogeneity was present. The certainty of the evidence will be ascertained through “GRADE (Grading of Recommendations, Assessment, Development, and Evaluation)”.
The SRMA can inform the framing of clinical guidelines for SCH management for all population groups and also intends to bring out gaps in the existing literature for future studies.
PROSPERO ID: CRD420251270021 (Date: 08 February 2026)
subclinical hypothyroidism, levothyroxine, pregnancy, adults, meta-analysis, evidence synthesis
Hypothyroidism is a major problem which affects 0.3 to 12% of the global population1 and contributes significantly to morbidity.2 Broadly, it includes overt, central, and subclinical hypothyroidism, depending on the varied combinations of thyroid function test (TFT) markers. “Subclinical hypothyroidism (SCH)” is identified when “serum-free thyroxine (FT4)” and “free triiodothyronine (FT3)” levels are normal, but “thyroid-stimulating hormone (TSH)” levels are elevated.3 The characteristic laboratory finding of subclinical hypothyroidism is an increased TSH level along with a normal T4 level.4 While it is typically regarded as an asymptomatic condition, some patients may exhibit non-specific symptoms that could indicate hypothyroidism. The prevalence of SCH in the general population is notably high (4% to 20%) and varies by sex and age. It is more common in individuals over 60 years of age, with a prevalence of approximately 15% in women and 8% in men.5 Of individuals with grade 1 subclinical hypothyroidism, 60% experienced a return of thyrotropin to normal levels within a five-year period. The yearly likelihood of advancing to overt hypothyroidism in these patients ranges from 2% to 4%, and is influenced by the presence of TPO antibodies. Among older adults (aged 65 and above), 46% of those with “grade 1 subclinical hypothyroidism” (TSH levels ranging from 4.5 to 6.9 mU/L) saw their thyrotropin normalize within two years, in contrast to only 7% of those with grade 2 subclinical hypothyroidism.6 SCH in pregnancy significantly progresses to overt hypothyroidism in the post-pregnancy period (within 5 years).7 In pregnant women, SCH has been reported to be associated with multiple “adverse pregnancy outcomes (APOs),” such as preeclampsia and preterm birth, among others.8,9 Higher risk for pregnancy loss, hypertensive disorders of pregnancy and placental abruption in SCH group has also been reported in the past meta-analyses.10,11
While the therapeutic decision for overt hypothyroidism is fairly straightforward, treatment of SCH, especially the timing of the initiation of therapy, has been a point of discussion. In pregnant women, levothyroxine (LT4) treatment has been shown to reduce APOs, such as preterm birth, gestational hypertension, pregnancy loss, and neonatal mortality.12–14 Effect of LT4 therapy in specific sub-groups such as recurrent pregnancy loss is a gap to be explored in evidence synthesis.15 Although multiple meta-analyses have reported the efficacy of LT4 therapy on pregnancy outcomes,16 relatively long-term outcomes such as neurodevelopment in infancy have not been reported. For adult SCH patients, levothyroxine treatment has been recommended since low-quality evidence was present for the critical outcomes (as of 2019).17 Peng In their systematic review of studies conducted until April 2020, Peng et al. reported a differential pattern of LT4 association with mortality among older and younger adults.18 Treatment for SCH also needs to factor in the TPOAb positive and levels of TSH at diagnosis, since these factors have been shown to increase the risk of progression of SCH to overt hypothyroidism.9,19 An updated pooled systematic review on the effect of LT4 among the adults (non-pregnant women, young adults and old adults) is not available. In addition, important effect modifiers such as age, sex, comorbidity, initial TSH levels of patients (as reported by the studies), symptom status, and the type of control group used (placebo/no treatment/watchful waiting/delayed initiation) need to be explored to ascertain any differential effects of LT4 therapy among patients with SCH.
Considering the current burden of SCH surpassing overt hypothyroidism, and that this hidden thyroid dysfunction can lead to complications, reduced quality of life, productivity, and lack of a comprehensive systematic review assessing the impact of LT4 therapy in SCH including all population groups, the current “systematic review and meta-analysis (SRMA)” is being proposed. The objective of this systematic review is to assess the effect of LT4 therapy on the prevention of complications, adverse events, and quality of life among SCH patients.
The SRMA protocol adhered to the “Preferred Reporting Items for Systematic reviews and Meta-Analyses Protocols (PRISMA-P 2015)” guidelines (Extended Data: Table S1). The review is registered in the PROSPERO database (CRD420251270021).
In individuals with subclinical hypothyroidism (P), what are the comparative effects of early initiation of treatment (I) compared to watchful waiting, delayed initiation, or no treatment (C) on progression to overt hypothyroidism and prevention of complications and their associated adverse outcomes (O)? The detailed eligibility criteria for the review are listed in Table 1.
RCTs will be included. Quasi-experimental, observational studies, reviews, case reports, case series, opinions, and commentaries will be excluded.
Studies conducted across the globe and published in English language will be included.
An independent systematic literature search will be performed to identify studies that evaluate the correct timing of treatment initiation. The databases “PubMed, Scopus, EMBASE, and the Cochrane Library” will be searched from inception until 30.12.2025. The search strategy was prepared by one of the reviewers and peer reviewed by another external reviewer by adhering to the PRESS guidelines20 (Extended Data: Table S2). Combination of “Medical Subject Headings (MeSH)” and free-text terms were used to frame the search strategy. Search strategies for the included databases are attached as Extended Data (Table S3).
A consensus meeting was conducted to finalize the search strategy. The records retrieved from the searches will be screened using predefined inclusion and exclusion criteria. Additional methods for identifying relevant studies, including manual screening of reference lists, within the eligible studies from the database search, will be undertaken.
Studies identified through the systematic search will be imported into the Nested Knowledge platform,21 where deduplication will be performed, followed by screening . The initial screening of the titles and abstracts will be performed by two independent reviewers.
Full-text assessment: Potentially eligible studies will undergo full-text retrieval and will be assessed independently by two reviewers to confirm eligibility for inclusion.
Disagreements: Any discrepancies in screening will be resolved through discussion between the two reviewers, and unresolved conflicts will be resolved by consulting a third reviewer.
The reference list of related systematic reviews identified during the screening process and the reference list of eligible primary studies will also be screened by two reviewers to identify any further eligible studies. Any discrepancies in screening will be resolved through discussion between the two reviewers, and unresolved conflicts will be resolved by consulting a third reviewer.
Two reviewers independently will extract the data using a process to resolve differences. A data extraction sheet will be constructed in MS Excel and piloted before the data extraction. Two independent reviewers will have a consensus meeting at the end of the independent data extraction to resolve differences. Any conflicts that could not be resolved will be adjudicated by a third reviewer.
Data extraction will include information on study identifiers, location, study population characteristics (age, sex, co-morbidities, pregnant/non-pregnancy, diagnosis methods, criteria for diagnosis), intervention group details (demography, dose, frequency, type of drug, duration, and any other relevant information as disclosed in the studies), comparator group details (demography, monitoring details, any other relevant information reported by the studies), outcomes such as progression from subclinical to overt hypothyroidism, prevention of disease-related complications (including maternal, fetal, neonatal, and other subgroup-specific outcomes), and adverse effects associated with LT4 therapy. Data on improvements in hypothyroidism-related symptoms, changes in health-related quality of life, and health system outcomes, such as cost-effectiveness, feasibility, acceptability, and equity, will also be extracted where reported.
The risk of bias for the “randomized controlled trials (RCTs)” will be evaluated through Cochrane’s ROB 2.0,22 independently by at least two reviewers, with a process to resolve differences. Two independent reviewers will have a consensus meeting at the end of the independent risk of bias assessment to resolve the differences. Any conflicts that could not be resolved will be adjudicated by a third reviewer.
Pooled estimates will be calculated for meta-analysis. “Risk ratio or Odds ratio or Hazard ratio” along with “95% confidence intervals (95% CI)” will be calculated for the dichotomous outcomes (e.g., progression to overt hypothyroidism, adverse events, maternal or neonatal outcomes). “Mean difference (MD)” or “standardized mean difference (SMD)” with 95% CI will be calculated for continuous outcomes such as symptom scores and quality-of-life measures. Pooled estimates for the outcomes will be calculated using a “random-effects model” (REM) with maximum likelihood estimation if the heterogeneity is high (I2>50%). Inter-study heterogeneity will be quantified using the I2 statistic. Sub-group analysis and meta-regression will be performed to explore heterogeneity, if present.23 Sub-group analysis based on TSH levels for diagnosis of SCH, age, life stage and reproductive status, co-morbidities, symptom status, autoimmune status, between different comparators, treatment duration, follow-up duration, environmental or geographic, or any other factor identified during the review will be undertaken. Sensitivity analyses will be performed using a leave-one-out approach to examine whether the overall results are unduly influenced by a single study. Publication bias assessment using funnel plots and the Eggers test index will be performed if more than 10 eligible studies are included in the meta-analysis. Statistical evaluations will be performed using RevMan Web.24 For outcomes where meta-analysis is not feasible, narrative synthesis will be provided.
“GRADE (Grading of Recommendations, Assessment, Development and Evaluation)” approach will be used for assessing the certainty (confidence) of the evidence for each outcome. The domains considered will include “risk of bias, inconsistency, indirectness, imprecision, and publication bias.” The GRADEpro tool will be used to generate a summary table of certainty in evidence.25
The decision to initiate hormone replacement therapy for SCH has been controversial, with varied evidence profiles in different population groups. For pregnant women, it has been recommended to initiate LT4 therapy in cases of SCH,19 whereas recommendations for adult SCH patients are against LT4 therapy.17 However, even within pregnancy outcomes, systematic reviews have reported differential pooled estimates, with favorable (reduced APOs),14 and neutral (no change in APOs),26 findings for LT4 therapy in SCH. The impact of co-morbidities on the effect of LT4 in SCH patients and specific outcomes, such as cardiovascular events, has been studied by certain primary studies.27,28 Studies have also assessed age group specific impact of the LT4 therapy in SCH outcomes, as well.28,29 Effect of the LT4 therapy in SCH infertile women on their fertility outcomes has also been reported in the past.30
The lack of certainty assessment (e.g., GRADE profile) in the outcomes reported is a lacuna in the existing systematic reviews on this topic,14,15,18,31 which seriously limits the transferability for guideline development. Through the current systematic review, we intend to provide a comprehensive, updated (until December 2025) evidence profile on the effect of LT4 therapy on important and critical patient outcomes (progression to overt hypothyroidism, mortality, complications, quality of life, acceptability, and costs) for multiple population groups (adults, non-pregnant women, infertile women, pregnant women, neonates, children, and co-morbidities). Subgroup analysis was planned in this review to explore the effect modification potential of all critical factors (age, TSH range for SCH diagnosis, symptom status, autoimmunity, and type of control group) on the direction and magnitude of the association between LT4 treatment and outcomes in SCH patients. GRADE profiling of critical and important outcomes will be performed in the index systematic review, which can provide certainty in the evidence pooled in the review.
Overall, the findings of the systematic review can inform the framing of the clinical guidelines for SCH management for all population groups, with or without conditional recommendations, as per the variations that may arise due to the subgroup analysis. The review can also reveal the differential effects, if any, according to the severity of the SCH (TSH levels and symptom status), which can enable layered recommendations instead of a blanket recommendation. The outputs from the SRMA can ensure robust guidelines centered on systematically collated, transparent, and critically appraised evidence, which in turn can enhance clinical practice and patient outcomes.
The SRMA also intends to bring out gaps in the existing literature (in terms of lack of adequate evidence in specific populations, specific outcomes, sub-groups, and quality of the studies conducted thus far). This can enable the formulation of research questions for future studies on the management of SCH using LT4 to address these gaps.
The authors declare that they have not used AI tools in the creation of this article.
Figshare: PRESS 2015 Guideline Evidence-Based Checklist for search strategy (Table S2) https://doi.org/10.6084/m9.figshare.31820230.v1.32
Figshare: Search strategies for the databases (Table S3) https://doi.org/10.6084/m9.figshare.31820242.v2.33
Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P 2015) (Table S1) https://doi.org/10.6084/m9.figshare.31820203.v1.34
Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).
Authors would like to thank the subject and methodological experts of the Technical Resource Center, Centre for Evidence Based Guidelines, Department of Community Medicine, AIIMS Nagpur, India, and the Technical Resource Hub, Center for Evidence Based Guidelines, DMIHER, Wardha, India, for reviewing the protocol and providing comments, which have helped in improving the quality.
| Views | Downloads | |
|---|---|---|
| F1000Research | - | - |
|
PubMed Central
Data from PMC are received and updated monthly.
|
- | - |
Provide sufficient details of any financial or non-financial competing interests to enable users to assess whether your comments might lead a reasonable person to question your impartiality. Consider the following examples, but note that this is not an exhaustive list:
Sign up for content alerts and receive a weekly or monthly email with all newly published articles
Already registered? Sign in
The email address should be the one you originally registered with F1000.
You registered with F1000 via Google, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Google account password, please click here.
You registered with F1000 via Facebook, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Facebook account password, please click here.
If your email address is registered with us, we will email you instructions to reset your password.
If you think you should have received this email but it has not arrived, please check your spam filters and/or contact for further assistance.
Comments on this article Comments (0)