Prevalence and incidence of hypothyroidism and hyperthyroidism in patients with chronic kidney disease on dialysis: a systematic review and meta-analysis

Abstract

Background and hypothesis

Chronic kidney disease (CKD) affects thyroid metabolism, increasing the risk of hypothyroidism and hyperthyroidism. Dialysis further alters hormone levels through impaired synthesis, iodine imbalance, and treatment-related factors. This study aimed to evaluate the prevalence and incidence of hypothyroidism and hyperthyroidism in patients with CKD on dialysis.

Methods

We conducted a search in the Scopus and PubMed databases up to November 29, 2024. We included studies reporting prevalence or incidence data on hypothyroidism or hyperthyroidism (clinical or subclinical) in CKD patients on dialysis (hemodialysis or peritoneal dialysis). A random-effects model meta-analysis of proportions was used to calculate pooled prevalence estimates.

Results

A total of 39 studies were included, with sample sizes ranging from 40 to 8840 participants. The prevalence of hypothyroidism was 15.9% (95% CI: 13.0 to 19.0; I²: 96.2%), with 5.6% for clinical hypothyroidism and 11.2% for subclinical hypothyroidism (notably higher in peritoneal dialysis). The prevalence of hyperthyroidism was 5.1% (95% CI: 2.7 to 8.2; I²: 93.8%), with 0.9% for clinical hyperthyroidism and 3.3% for subclinical hyperthyroidism. Meta-regression analyses indicated that a longer mean duration on dialysis was associated with a lower prevalence of hypothyroidism (p = 0.008). Only one study reported the incidence of hypothyroidism (10.9%) and hyperthyroidism (4.9%), thus evidence on incidence remains scarce.

Conclusion

The prevalence of hypothyroidism and hyperthyroidism in patients with CKD on dialysis is high, although with high heterogeneity among studies. Evidence on incidence remains limited and should be interpreted as exploratory. These findings support increased clinical vigilance and consideration of early detection strategies rather than definitive screening policies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12882-026-04975-1.

Background

Chronic kidney disease (CKD) has a global prevalence of approximately 8% and represents a significant cause of mortality [1]. In recent years, the incidence of this disease, mortality rates, and disability-adjusted life years have significantly increased [2]. Stage 5 CKD affects approximately 0.1% of the population, and many patients begin dialysis treatment, which is available in most countries [3, 4]. Hemodialysis is the predominant modality, accounting for approximately 88% of cases, while peritoneal dialysis is used in the remaining 12% [5]. These patients face a high risk of infections at the access site, which are the leading cause of hospitalization, while cardiovascular disease remains the primary cause of death [4].

In addition to these complications, CKD is associated with profound alterations in thyroid hormone metabolism. In the general population, hypothyroidism or hyperthyroidism affect approximately 0.2 to 5.3% and 0.2 to 1.3%, respectively [6]. In patients with advanced CKD and those undergoing dialysis, thyroid dysfunction is more plausibly explained by impaired renal clearance of iodine, reduced peripheral conversion of T4 to T3, chronic inflammation, and dialysis-related metabolic disturbances [7–9]. Also, iodine excess, certain treatments, or heparin used during hemodialysis has been associated with hyperthyroid biochemical patterns in some settings [10]. Previous primary studies have documented a higher prevalence of clinical or subclinical thyroid disorders in patients undergoing hemodialysis or peritoneal dialysis [11, 12]. These disorders can worsen renal deterioration and increase cardiovascular mortality in dialysis patients [13, 14].

A systematic review (search period: 2013 to 2023) assessed the prevalence of thyroid disorders in CKD patients; however, it included a limited number of studies focusing on dialysis patients and did not conduct a meta-analysis [15]. To date, no systematic review has specifically quantified the prevalence of hypothyroidism and hyperthyroidism in patients with CKD undergoing hemodialysis or peritoneal dialysis. Clarifying these prevalences may support improved clinical awareness, inform treatment considerations, and contribute to better management and quality of life in this population.

Therefore, the objective of this systematic review was to evaluate the prevalence and incidence of hypothyroidism and hyperthyroidism in patients with CKD on dialysis.

Materials and methodology

We conducted a systematic review following the guidelines recommended in the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Items 15 and 22 were deemed not applicable because no GRADE guideline currently standardizes the assessment of certainty of evidence in meta-analyses of proportions. The protocol for this review has been registered in PROSPERO under the number CRD42025622678. A modification to the protocol was made to include the analysis of both prevalence and incidence.

Selection criteria

We included cross-sectional and cohort studies reporting the prevalence or incidence of hypothyroidism and hyperthyroidism (clinical or subclinical) in CKD patients on dialysis (hemodialysis or peritoneal dialysis) in any setting. Exclusion criteria were duplicate populations, clinical trials, case-control studies, case reports, editorials, commentaries, clinical practice guidelines, reviews, manuscripts not available in full text, or studies with fewer than 30 participants.

Search strategy & sources

A systematic advanced search was conducted in the Scopus and PubMed/MEDLINE databases on November 29, 2024, without restrictions on language or publication date. The full search strategy is presented in S1 Table. Additionally, a manual search of reference lists from included studies was performed to identify other eligible studies.

Study selection

Study selection was carried out using the Rayyan software. Duplicate articles were manually removed. Authors (SDCD, JDCD, VMRV) independently screened titles and abstracts to identify potentially relevant studies. Selected studies were then evaluated in full text for inclusion independently by the same authors. Discrepancies were resolved through discussion among the authors, with a third reviewer (DRSM).

Data extraction

Authors (SDCD, JDCD, VMRV) independently extracted the following key data into Microsoft Excel software: author, year of publication, country, sample size, dialysis type, age and sex of participants, definition of thyroid disorder, mean dialysis duration, number of patients undergoing hemodialysis and peritoneal dialysis, and cases of clinical and subclinical hypothyroidism and hyperthyroidism by dialysis type. Any discrepancies between the authors were resolved through discussion and consensus, with the involvement of a more experienced third author (DRSM). In cases of duplicate populations, the study providing the most relevant information was included.

Risk of bias assessment

The authors (SDCD, JDCD, VMRV) independently assessed the risk of bias in the included studies using the Joanna Briggs Institute Critical Appraisal Tool for prevalence and incidence studies [16]. This tool consists of nine items, with possible responses of “yes,” “no”, or “unclear”. The scale score was calculated by assigning 1 point for meeting the criterion and 0 points if not met or unclear. Higher scores indicated better methodological quality. Study-level risk of bias was classified as low (7 to 9 points), moderate (4 to 6 points), or high risk of bias (0 to 3 points). Any discrepancies between the authors were resolved through discussion and consensus, with the involvement of a more experienced third author (DRSM).

Statistical analysis

Statistical analysis was performed using STATA V17.0. To calculate the prevalence of the primary outcomes, we used the sum of clinical and subclinical cases of hypothyroidism or hyperthyroidism as the numerator and the total number of dialysis patients (hemodialysis or peritoneal dialysis) as the denominator. Additionally, prevalence was calculated based on thyroid disorder type (subclinical hypothyroidism, clinical hypothyroidism, subclinical hyperthyroidism, and clinical hyperthyroidism) and dialysis type (hemodialysis and peritoneal dialysis). Prevalence estimates with 95% confidence intervals (95% CI) were calculated using meta-analysis of proportions with a random-effects model. The 95% CIs were computed using the exact method, and variances were stabilized by applying the Freeman-Tukey Double Arcsine transformation [17, 18]. In addition, 95% prediction intervals (95% PI) were calculated to estimate the expected range of prevalence values for future studies. Heterogeneity was assessed using the I² statistic, and subgroup analyses were conducted to explore sources of heterogeneity by country, year of publication, age group, risk of bias and thyroid disorder definition [6, 19]. Furthermore, meta-regression analyses were performed based on mean age, year of publication, percentage of male participants, average dialysis duration, and risk of bias score. Publication bias was evaluated using a funnel plot for asymmetries and Egger’s test, considering a P-value < 0.05 as indicative of publication bias.

Results

Study selection

In the systematic search, we identified 2330 studies after removing duplicates, which were screened by title and abstract. A total of 254 studies were selected for full-text evaluation. Ultimately, 39 studies were included in this systematic review (Fig. 1). The excluded studies at the full-text stage, along with the reasons for exclusion, are presented in S2 Table.

Fig. 1.

Flowchart summarizing the literature search and selection process

Characteristics of included studies

The 39 included studies were conducted in various countries, including Germany [20], Saudi Arabia [12, 21, 22], Bosnia and Herzegovina [23], Brazil [24], China [25], Korea [26, 27], Egypt [28], United States [9, 29–33], Greece [34], India [35, 36], Iraq [37], Iran [38, 39], Japan [40], Morocco [41], Mexico [42, 43], Nepal [44], Pakistan [45, 46], Palestine [11], Czech Republic [47], Romania [48], Russia [49], Serbia [50], Somalia [51], Taiwan [52–54] and Turkey [55]. Regarding dialysis type, most studies focused on hemodialysis [9, 20, 21, 23, 24, 28, 30–32, 34–41, 44–53, 55], while some included patients undergoing peritoneal dialysis [26, 27, 54] or both treatments [11, 12, 22, 25, 33, 43]. Sample sizes ranged from 40 to 8840 participants. The mean age of subjects varied between 13.2 and 67.5 years, the proportion of men ranged from 29.3% to 78.1%, and the average dialysis duration spanned from 0.7 to 9.4 years. Regarding specific populations, two studies specifically included patients receiving erythropoietin therapy [53, 54], one study focused on patients with a life expectancy of less than six months [30] and one study included pediatric patients aged 4 to 17 years [43]. The studies define hypothyroidism as TSH levels above 4.5 to 10 mIU/L, distinguishing clinical hypothyroidism by low FT4 and subclinical hypothyroidism by normal FT4. The studies defined hyperthyroidism as TSH levels below 0.01 to 0.5 mIU/L, where clinical cases exhibit elevated FT4 and/or FT3, while subclinical cases retain normal thyroid hormone levels. The prevalence of hypothyroidism ranged from 1.8 to 30.6%, while that of hyperthyroidism varied from 0.7 to 14.7% (Table 1).

Table 1.

Characteristics of the included studies evaluating the prevalence of hypothyroidism and hyperthyroidism in patients with chronic kidney disease on dialysis (n = 39)

Author-year	Country	Dialysis Type	Sample Size	Age (years) Mean ± SD	Male (%)	Dialysis Duration (years) Mean ± SD	Prevalence of Hypothyroidism	Prevalence of Hyperthyroidism	Risk of Bias Score (Max. 9)
Rhee-2016	USA	PD	1484	60.0 ± 15.0	48.0	NR	18.2	7.0	8
Pakfetrat-2017	Iran	HD	86	57.2 ± 17.2	55.8	2.4 ± 0.5	18.6	NR	9
Rhee-2017	USA	HD	541	54.0 ± 15.0	55.1	4.4 ± 1.1	10.5	3.3	9
Da-2016	Brazil	HD	58	47.4 ± 12.3	41.4	4.9 ± 3.6	NR	NR	7
Rhee-2015	USA	HD	8840	65.0 ± 15.0	51.0	2.0 ± 2.0	21.8	NR	8
Naseem-2018	Pakistan	HD	72	36.3 ± 9.4	63.9	NR	NR	NR	6
Lin-2018	Taiwan	HD	4540	NR	29.3	NR	20.0	NR	5
Kaptein-1988	USA	HD	248	NR	NR	NR	NR	NR	7
Ng-2013	Taiwan	HD	1013	61.3 ± 13.4	48.9	6.8 ± 6.6	9.7	5.0	8
Kutlay-2005	Turkey	HD	87	42.9 ± 11.8	54.0	4.5 ± 3.8	NR	NR	8
Horáček-2012	Czech Republic	HD	167	64.9 ± 14.9	58.1	NR	11.4	3.6	8
Abdullah-2023	Pakistan	HD	200	46.2 ± 19.6	56.5	NR	NR	NR	7
Najoua-2010	Morocco	HD	68	43.4 ± 13.5	38.2	9.4 ± 6.6	27.9	NR	4
Adani-2023	Somalia	HD	301	54.0 ± 19.3	55.5	NR	27.6	1.7	9
Ahmad-2021	India	HD	89	57.6 ± 14.5	NR	NR	16.9	NR	6
Sinjari-2022	Iraq	HD	104	53.9 ± 14.6	51.9	4.7 ± 3.6	27.9	4.8	8
Bichari-2020	Egypt	HD	100	56.9 ± 15.3	57.0	6.1 ± 1.0	7.0	NR	7
Eskandarifar-2022	Iran	HD	108	61.1 ± 13.5	53.7	NR	7.4	NR	9
Kovalevsky-2020	Russia	HD	108	51.7 ± 15.4	59.3	NR	NR	NR	7
Cotoi-2020	Romania	HD	123	62.2 ± 11.0	56.9	5.6 ± 4.8	24.4	3.3	9
Bin-2018	Saudi Arabia	HD	130	57.1 ± 19.2	46.9	NR	27.7	NR	8
Shantha-2011	South India	HD	137	43.0 ± 13.3	78.1	5.6 ± 2.2	NR	NR	8
Lo-2017	USA	HD	226	NR	NR	NR	11.1	NR	5
Al-2023	Saudi Arabia	PD and HD	99	51.3 ± 16.9	76.8	3.8 ± 3.9	9.1	NR	6
Jusufovic-2011	Bosnia and Herzegovina	HD	40	53.2 ± 11.7	45.0	5.0 ± 3.0	22.5	NR	6
Nazzal-2020	Palestine	PD and HD	209	57.6 ± 14.5	64.1	3.0 ± 0	NR	NR	8
Alshammari-2019	Saudi Arabia	PD and HD	255	NR	62.7	NR	20.4	NR	8
Sanai-2015	Japan	HD	145	60.8 ± 14.5	49.7	6.7 ± 5.9	11.7	NR	8
Drechsler-2014	Germany	HD	1000	65.6 ± 8.5	NR	NR	1.8	14.7	6
Paudel-2014	Nepal	HD	64	47.2 ± 15.6	76.6	NR	26.6	NR	7
Rhee-2013	USA	PD and HD	2715	63.3 ± 15.5	55.8	NR	12.9	NR	9
Lazarevic-2015	Serbia	HD	50	58.8 ± 10.3	66.0	5.6 ± 5.0	8.0	NR	6
Jung-2014	Korea	PD	235	51.4 ± 13.3	56.2	5.4 ± 4.3	NR	NR	5
Pámanes-2014	Mexico	PD	76	58.3 ± 12.3	56.6	2.3 ± 1.4	26.3	NR	6
Ng-2012	Taiwan	PD	122	NR	NR	NR	NR	NR	8
Garrido-Magaña-2009	Mexico	PD and HD	50	13.2 ± 9.1	50.0	NR	22.0	NR	8
Kang-2008	Korea	PD	51	63.0 ± 38.1	51.0	0.7 ± 3.8	NR	NR	7
Kalocheretis-2004	Greece	HD	109	67.5 ± 8.8	56.0	4.8 ± 4.3	11.0	NR	6
Lin-1998	China	PD and HD	221	57.0 ± 15.2	56.6	3.8 ± 3.4	NR	NR	9

NR: Not reported; SD: Standard Deviation; PD: Peritoneal Dialysis; HD Hemodialysis; USA: United States of America

Regarding the incidence of thyroid disorders, only one study, conducted in the United States with 1484 peritoneal dialysis patients, reported a hypothyroidism incidence of 10.9% (TSH ≥ 5.0 mIU/L) and a hyperthyroidism incidence of 4.9% (TSH < 0.5 mIU/L) (S3 Table).

Risk of bias

Most studies met criteria for appropriate statistical analysis (100%), adequate response rate (94.9%), condition measured in a standard, reliable way for all participants (94.9%), and adequate sampling framework (92.3%). However, deficiencies were identified in valid methods used for the identification of the condition (61.5%), sufficient coverage of the identified sample (59.0%) and sample size adequate (56.4%). Overall scores ranged from 4 to 9, with a median of 7. The general evaluation scores are presented in Table 1 and detailed in S4 Table.

Prevalence of hypothyroidism

The combined prevalence of clinical or subclinical hypothyroidism in patients with CKD on dialysis was 15.9% (95% CI: 13.0 to 19.0; 95% PI: 4.0 to 33.5; I²: 96.2%), with a range of 7.0 to 27.9% (Fig. 2). The prevalence of clinical hypothyroidism was 5.6% (95% CI: 3.7 to 7.8; 95% PI: 0.1 to 18.7; I²: 94.7%), and that of subclinical hypothyroidism was 11.2% (95% CI: 9.0 to 13.6; 95% PI: 2.3 to 25.2; I²: 92.5%) (S1 and S2 Figs).

Fig. 2.

Prevalence of clinical or subclinical hypothyroidism in patients on any type of dialysis

In hemodialysis patients, hypothyroidism prevalence was 15.7% (95% CI: 12.3 to 19.5; 95% PI: 2.6 to 36.4; I²: 96.6%), while in peritoneal dialysis patients, it reached 25.0% (95% CI: 14.4 to 37.2; 95% PI: 0.1 to 69.6; I²: 82.7%) (S3 and S4 Figs, Table 2).

Table 2.

Prevalence of clinical and subclinical hypothyroidism and hyperthyroidism in hemodialysis and peritoneal dialysis

	Any Dialysis				HD				PD
Thyroid Disorder	n studies	Prevalence (%)	95% CI	I2 (%)	n studies	Prevalence (%)	95% CI	I2 (%)	n studies	Prevalence (%)	95% CI	I2 (%)
Hypothyroidism	27	15.9	13.0 to 19.0	96.2	23	15.7	12.3 to 19.5	96.6	5	25.0	14.4 to 37.2	82.7
Clinical Hypothyroidism	22	5.6	3.7 to 7.8	94.7	17	5.8	3.2 to 9.1	95.6	3	10.1	1.3 to 24.5	78.4
Subclinical Hypothyroidism	28	11.2	9.0 to 13.6	92.5	21	10.4	7.5 to 13.6	93.7	7	18.3	14.3 to 22.7	25.9
Hyperthyroidism	8	5.1	2.7 to 8.2	93.8	7	4.8	1.9 to 8.8	94.6	1	7.0	5.8 to 8.4
Clinical Hyperthyroidism	8	0.9	0.3 to 1.8	64.9	7	0.8	0.2 to 1.9	67.7
Subclinical Hyperthyroidism	12	3.3	1.3 to 6.0	90.9	9	3.7	1.3 to 7.2	91.2	2	2.7	1.2 to 4.7	0.0

95% CI: 95% confidence interval; PD: Peritoneal Dialysis; HD Hemodialysis

Prevalence of hyperthyroidism

The combined prevalence of clinical or subclinical hyperthyroidism in patients with CKD on dialysis was 5.1% (95% CI: 2.7 to 8.2; 95% PI: 0.1 to 18.8; I²: 93.8%) (Fig. 3). The prevalence of clinical hyperthyroidism was 0.9% (95% CI: 0.3 to 1.8; 95% PI: 0.1 to 4.3; I²: 64.9%), and that of subclinical hyperthyroidism was 3.3% (95% CI: 1.3 to 6.0; 95% PI: 0.1 to 17.3; I²: 90.9%) (S5 and S6 Figs).

Fig. 3.

Prevalence of clinical or subclinical hyperthyroidism in patients on any type of dialysis

In hemodialysis patients, hyperthyroidism prevalence was 4.8% (95% CI: 1.9 to 8.8; 95% PI: 0.1 to 23.7; I²: 94.6%), while in peritoneal dialysis patients, it was 7.0% (95% CI: 5.8 to 8.4%) (S7 and S8 Figs, Table 2).

Subgroup analysis

We evaluated the prevalence of hypothyroidism and hyperthyroidism in CKD patients on dialysis according to country, year of publication, age, and risk of bias. Hypothyroidism prevalence was highest in Morocco (27.9%) and Iraq (27.9%), while the lowest were found in Germany (1.8%) and Egypt (7.0%) (S9 Fig). Hyperthyroidism prevalence was highest in Germany (14.7%) and lowest in Somalia (1.7%) (S10 Fig).

Prevalence estimates remained similar across publication years for both hypothyroidism and hyperthyroidism (S11 and S12 Figs).

Regarding age, only one study included pediatric patients, reporting a hypothyroidism prevalence of 22.0%, while in adults, the prevalence was 15.7% (S13 Fig). The pediatric study did not report hyperthyroidism prevalence, so it was only evaluated in adults, where it was 5.1% (S14 Fig, Table 3).

Table 3.

Prevalence of hypothyroidism and hyperthyroidism and subgroups in patients with chronic kidney disease on dialysis

	Hypothyroidism				Hyperthyroidism
	n studies	Prevalence (%)	95% CI	I2 (%)	n studies	Prevalence (%)	95% CI	I2 (%)
Country
USA	5	14.8	10.5 to 19.8	97.5	2	5.9	4.9 to 7.0
Iran	2	11.8	7.6 to 16.8
Taiwan	2	17.9	16.9 to 18.9		1	5.0	3.8 to 6.6
Czech Republic	1	11.4	7.0 to 17.2		1	3.6	1.3 to 7.7
Morocco	1	27.9	17.7 to 40.1
Somalia	1	27.6	22.6 to 33.0		1	1.7	0.5 to 3.8
India	1	16.9	9.8 to 26.3
Iraq	1	27.9	19.5 to 37.5		1	4.8	1.6 to 10.9
Egypt	1	7.0	2.9 to 13.9
Romania	1	24.4	17.1 to 33.0		1	3.3	0.9 to 8.1
Saudi Arabia	3	18.6	10.1 to 29.0
Bosnia and Herzegovina	1	22.5	10.8 to 38.5
Japan	1	11.7	7.0 to 18.1
Germany	1	1.8	1.1 to 2.8		1	14.7	12.6 to 17.0
Nepal	1	26.6	16.3 to 39.1
Serbia	1	8.0	2.2 to 19.2
Mexico	2	24.6	17.3 to 32.6
Greece	1	11.0	5.8 to 18.4
Year of publication
1988–2009	2	14.0	9.0 to 20.0
2010–2019	18	15.7	12.3 to 19.5	97.3	5	6.4	3.2 to 10.5	95.4
2020–2023	7	16.4	9.7 to 24.4	88.5	3	2.7	1.1 to 5.0
Age
Adult	26	15.7	12.8 to 18.9	96.4	8	5.1	2.7 to 8.2	93.8
Pediatric	1	22.0	11.5 to 36.0
Risk of bias
Low risk	17	17.1	14.0 to 20.4	94.4	7	4.1	2.7 to 5.7	76.2
Moderate risk	10	14.1	6.9 to 23.2	97.6	1	14.7	12.6 to 17.0

95% CI: 95% confidence interval; USA: United States of America

The prevalence of hypothyroidism, categorized by risk of bias, were similar between studies with a low risk of bias (17.1%) and those with a moderate risk of bias (14.1%) (S15 Fig). Similarly, for hyperthyroidism, prevalence estimates were comparable in low-risk studies (4.1%) and moderate-risk studies (5.1%) (S16 Fig). The subgroup analyses did not explain the heterogeneity observed among included studies.

The subgroup analysis according to TSH cut-off thresholds showed no clear gradient or consistent difference in the pooled prevalence of thyroid dysfunction across definitions. Although studies were stratified by diagnostic criteria (TSH 4–5, 5–9.9, and ≥ 10 mIU/L), the prevalence estimates largely overlapped, and no meaningful trend was observed between higher or lower thresholds. Likewise, the results were similar when stratified by dialysis modality (hemodialysis versus peritoneal dialysis) (S5 Table).

Meta-regression

In the meta-regression analyses, we identified that a longer average dialysis duration was associated with a lower prevalence of hypothyroidism (β = -0.021; 95% CI: -0.035 to -0.007; p = 0.008). In contrast, mean age, publication year, male percentage, and risk of bias score were not significantly associated with prevalence (Table 4).

Table 4.

Meta-regression analyses of factors associated with the prevalence of hypothyroidism in patients with chronic kidney disease on dialysis

Variable	n studies	β	95% CI	p value	R 2 Adjusted
Mean age (years)	24	-0.004	-0.009 a 0.001	0.152	6.6
Year of publication	27	0.006	-0.003 a 0.015	0.190	17.3
Male (%)	24	-0.001	-0.004 a 0.002	0.358	-1.0
Mean dialysis duration (years)	14	-0.021	-0.035 a -0.007	0.008	92.6
Risk of bias score	27	0.004	-0.020 a 0.029	0.681	-7.7

95% CI: 95% confidence interval

Publication bias

In the meta-analysis of hypothyroidism in CKD patients on dialysis, the funnel plot showed a symmetrical distribution. Additionally, Egger’s test was not significant (p = 0.259), suggesting no presence of publication bias (S17 Fig).

Discussion

Main findings

In this systematic review and meta-analysis, we found that the prevalence of hypothyroidism in patients with CKD on dialysis was 15.9% (5.6% clinical hypothyroidism and 11.2% subclinical). In contrast, hyperthyroidism was less frequent, with a prevalence of 5.1% (0.9% clinical hyperthyroidism and 3.3% subclinical). Prevalence estimates were similar between dialysis types, except for subclinical hypothyroidism, which was higher in peritoneal dialysis compared to hemodialysis. Additionally, prevalence remained similar across subgroups based on country, publication year, age, and risk of bias. The meta-regression analysis found that a longer dialysis duration was associated with a lower prevalence of hypothyroidism. Only one study assessed the incidence of hypothyroidism (10.9%) and hyperthyroidism (4.9%), therefore, incidence findings should be interpreted as exploratory.

Hypothyroidism in dialysis patients

The prevalence of hypothyroidism and hyperthyroidism in patients with CKD on dialysis is higher than that reported in the general population, which ranges from 0.2% to 5.3% for hypothyroidism and 0.2% to 1.3% for hyperthyroidism [6]. Comparing our results with previous systematic reviews, we found that the prevalence of thyroid disorders in CKD patients on dialysis is higher than that observed in conditions such as vitiligo [56] or COVID-19 [57], but lower compared to patients with diabetes [58] or chronic obstructive pulmonary disease [59]. These findings may be explained by altered T4 to T3 conversion, iodine accumulation, chronic inflammation, and hormonal loss during dialysis, all of which may be associated with hypothyroidism [8]. In addition, hypothyroidism prevalence may be underestimated because some patients are already receiving levothyroxine therapy, which can normalize TSH levels.

Hyperthyroidism in dialysis patients

Hyperthyroidism was less frequent in patients with CKD on dialysis [60]. This lower prevalence may be explained by reduced renal clearance of thyroid hormones, alterations in hypothalamic–pituitary–thyroid axis regulation, and regional differences in iodine intake [10, 61].

Subclinical hypothyroidism and clinical relevance

Among the thyroid disorders evaluated, subclinical hypothyroidism was the most frequent, with a prevalence of 11.2%, a finding consistent with previous reviews [56, 58]. Previous studies have shown that as the glomerular filtration rate decreases, the prevalence of subclinical hypothyroidism increases, possibly due to reduced iodine clearance, leading to elevated serum iodine levels and inhibition of thyroid hormone production [7]. Additionally, this disorder is particularly common in patients undergoing peritoneal dialysis [54]. Despite being subclinical, this condition is clinically relevant as it can affect systolic and diastolic cardiac function, endothelial function, blood pressure, and lipid profile [62]. Moreover, these patients may potentially benefit from thyroid hormone replacement therapy [63].

Heterogeneity

A high degree of heterogeneity was observed in the prevalence of thyroid disorders among patients with CKD undergoing dialysis. The subgroup analyses failed to fully explain this variability, although lower heterogeneity was noted in analyses of subclinical hypothyroidism and subclinical hyperthyroidism among patients receiving peritoneal dialysis. Likewise, the subgroup analysis based on TSH cut-off thresholds used to define thyroid dysfunction did not reveal a clear gradient or consistent trend across diagnostic definitions.

Among the potential sources of variability, dialysis modality plays an important role. Peritoneal dialysis may promote T₄ loss and increase iodine absorption due to exposure to iodine-containing disinfectants [25]. In contrast, hemodialysis tends to remove circulating iodine, producing opposite effects on thyroid function [11, 23]. Additionally, meta-regression analysis indicated that a longer duration of dialysis might be associated with a lower prevalence of hypothyroidism, possibly reflecting an adaptive physiological response to prolonged treatment, although the underlying mechanism remains unclear and may partly reflect survival bias or residual confounding rather than a true protective effect.

Beyond dialysis-related factors, part of the variability likely arises from methodological and clinical differences among studies. Variations in the timing of blood sampling (before or after dialysis) and the use of heparin during hemodialysis can transiently alter FT₄ and TSH levels, leading to artificially elevated or reduced measurements [12, 23, 47, 50]. Similarly, the use of different laboratory techniques-such as radioimmunoassay, electrochemiluminescence immunoassay, or immunoradiometric assay-can affect measurement precision and limit comparability across studies [12, 34, 40, 47]. Furthermore, differences in demographic and clinical characteristics, including comorbidities, inflammatory and nutritional status, iodine exposure, thyroid autoimmunity, and socioeconomic conditions, may further contribute to the observed variability across countries [24, 34, 40].

Overall, subgroup and meta-regression findings should be interpreted cautiously, given the exploratory nature of multiple analyses and the inherent risk of false-positive associations.

Limitations of included studies

The studies included in this review present several limitations. First, most data come from a limited number of countries, reducing their global representativeness. Among the 39 studies analyzed, only a few provided detailed information on the studied population, including dialysis duration and type, as well as associated risk factors. Also, evidence on the incidence of thyroid disorders was very scarce (only one study), which limits the precision of our conclusions and underscores the need for future cohort studies. Regarding risk of bias, most studies were rated as having low to moderate risk of bias, however, recurrent methodological limitations (particularly diagnostic variability, differing thyroid function thresholds, and limited sample coverage) may have influenced pooled prevalence estimates and contributed to the observed heterogeneity.

Implications and recommendations

These findings highlight the importance of timely diagnosis and treatment of thyroid disorders in these patients, particularly those undergoing peritoneal dialysis, where subclinical hypothyroidism is more prevalent. The absence of adequate treatment may worsen their clinical condition and increase the risk of cardiovascular and renal complications [13, 14]. Therefore, it is essential to implement early detection strategies and appropriate management to improve prognosis and quality of life in these patients.

Future research should standardize diagnostic criteria for hypothyroidism and hyperthyroidism, distinguishing between clinical and subclinical forms. Additionally, cohort studies reporting the incidence of thyroid disorders in CKD patients are needed, as evidence on this topic remains scarce.

Although thyroid dysfunction is common in patients with CKD on dialysis, these findings should not be interpreted as supporting universal screening. Given study heterogeneity and the observational nature of the evidence, results should inform clinical awareness and individualized assessment rather than definitive screening recommendations.

Strengths and limitations

This systematic review has several strengths. First, we conducted searches in multiple databases using advanced search strategies without language or date restrictions. All review processes were performed independently to ensure data quality. Additionally, we applied subgroup and meta-regression analyses to assess potential sources of heterogeneity. However, despite the broad coverage of PubMed and Scopus, some regional studies indexed in other databases may have been missed. In addition, studies from low-resource regions were under-represented, which may limit the global generalizability of the findings. Another limitation is that we did not include gray literature to complement our search. In this context, although we formally assessed publication bias, it is important to recognize that conventional funnel plots may be unreliable in meta-analyses of proportions, particularly when outcomes are rare or extreme [64]. Finally, non-thyroidal illness syndrome—a common biochemical pattern in patients with advanced CKD characterized by low T3 with normal TSH—was not evaluated in this review, as it does not represent primary thyroid dysfunction.

Conclusion

The prevalence of hypothyroidism and hyperthyroidism is high in CKD patients on dialysis, although with significant heterogeneity among studies. Subclinical hypothyroidism was more frequent in peritoneal dialysis patients, whereas other subgroups showed similar prevalence estimates. Evidence on incidence remains limited and should be interpreted as exploratory. These findings support increased clinical vigilance and consideration of early detection strategies rather than definitive screening policies. Future high-quality prospective cohort studies are needed to better define incidence and inform clinical decision-making.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CKD

Chronic Kidney Disease

CI

Confidence Interval

PI

Prediction interval

FT3

Free Triiodothyronine

FT4

Free Thyroxine

T3

Triiodothyronine

T4

Thyroxine

TSH

Thyroid-Stimulating Hormone

Author contributions

Conceptualization: Sharong D. Castro-Diaz, Jennifer D. Castro-Diaz, Viviana M. Ruiz-Vargas Data curation: Sharong D. Castro-Diaz, Jennifer D. Castro-Diaz, Viviana M. Ruiz-Vargas Formal analysis: David R. Soriano-Moreno Investigation: All authors Methodology: All authors Project administration: Anderson N. Soriano-Moreno, David R. Soriano-Moreno Resources: All authors Software: David R. Soriano-Moreno Supervision: Anderson N. Soriano-Moreno, David R. Soriano-Moreno Writing – original draft: All authors Writing – review & editing: All authors.

Funding

This research received no external funding.

Data availability

All data generated or analysed during this study are included in this published article and its supplementary information files.

Declarations

Ethics approval and consent to participate

This study is a systematic review that used data extracted exclusively from previously published original articles and did not involve the collection of new data from human participants. Therefore, ethical approval and informed consent were not required. Nonetheless, the study protocol was reviewed and approved by the Ethics and Bioethics Committee of the Faculty of Health Sciences at Universidad Peruana Unión, under approval number 2024-CEB-FCS-UpeU-«N°215», granted on September 17, 2024, and valid until September 17, 2025.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.