Heart rate variability in hypothyroid patients: A systematic review and meta-analysis.

INTRODUCTION: Hypothyroidism may be associated with changes in the autonomic regulation of the cardiovascular system, which may have clinical implications.
OBJECTIVE: To conduct a systematic review and meta-analysis on the impact of hypothyroidism on HRV.
MATERIALS AND METHODS: PubMed, Cochrane, Embase and Google Scholar were searched until 20 August 2021 for articles reporting HRV parameters in untreated hypothyroidism and healthy controls. Random-effects meta-analysis were stratified by degree of hypothyroidism for each HRV parameters: RR intervals (or normal to normal-NN intervals), SDNN (standard deviation of RR intervals), RMSSD (square root of the mean difference of successive RR intervals), pNN50 (percentage of RR intervals with >50ms variation), total power (TP), LFnu (low-frequency normalized unit), HFnu (high-frequency), VLF (very low frequency), and LF/HF ratio.
RESULTS: We included 17 studies with 11438 patients: 1163 hypothyroid patients and 10275 healthy controls. There was a decrease in SDNN (effect size = -1.27, 95% CI -1.72 to -0.83), RMSSD (-1.66, -2.32 to -1.00), pNN50 (-1.41, -1.98 to -0.84), TP (-1.55, -2.1 to -1.00), HFnu (-1.21, -1.78 to -0.63) with an increase in LFnu (1.14, 0.63 to 1.66) and LF/HF ratio (1.26, 0.71 to 1.81) (p <0.001). HRV alteration increased with severity of hypothyroidism.
CONCLUSIONS: Hypothyroidism is associated with a decreased HRV, that may be explained by molecular mechanisms involving catecholamines and by the effect of TSH on HRV. The increased sympathetic and decreased parasympathetic activity may have clinical implications.

Introduction

The heart is richly innervated by vagal and sympathetic fibers and is sensitive to autonomic influences [1]. The autonomic nervous system, by its sympathetic and parasympathetic divisions, regulates and modulates involuntary body functions. Dysautonomia refers to a change in the function of the autonomic nervous system that negatively affects a person’s health [2], including increased cardiovascular morbidity [3]. Thyroid insufficiency or hypothyroidism is the inability of the thyroid gland to produce enough thyroid hormone. It is the most common hormonal disorder with a prevalence of 4–9% in women and 1–3% in men [4, 5]. Clinical signs of hypothyroidism include cardiovascular signs (bradycardia, decreased cardiac output and cardiac contractility] and suggest hypoactivity of the sympathetic nervous system [6]. If undiagnosed or insufficiently supplemented, hypothyroidism may be associated with changes in the autonomic regulation of the cardiovascular system. Heart rate variability (HRV) consists of the measurement of the physiological variation of RR intervals, a simple and convincing diagnostic tool used to assess the cardiac component of the autonomic nervous system [7-10]. Low HRV is an independent predictor of cardiac morbidity [11], while high HRV suggests good ability to adapt and respond to internal and external stimuli [3, 12]. Many studies have evaluated HRV parameters in hypothyroidism, but the results remain contradictory [13-18], although all tend to express the existence of alterations in parasympathetic and sympathetic activities in hypothyroidism compared with healthy controls. Few studies have comprehensively evaluated the role of the most common variables, such as age, sex, body mass index (BMI), blood pressure or biochemical thyroid function on HRV parameters in hypothyroidism [19, 20]. Therefore, we aimed to conduct a systematic review and meta-analysis on the impact of untreated hypothyroidism on HRV parameters. A secondary objective was to identify the most frequently reported predictors.

Methods

Literature search

All studies measuring HRV in patients with untreated hypothyroidism and healthy controls were reviewed until August 20, 2021, on the major article databases (PubMed, Cochrane Library, Embase, and Google Scholar) with the following keywords: ("hypothyroidism" OR "hypothyroid") AND ("heart rate variability" OR "HRV"). We included all articles that met our inclusion criteria of measuring HRV parameters in hypothyroid patients and healthy controls, regardless of article language and year of publication. There were no restrictions on the regional origin or nature of the control group. We excluded studies evaluating the effect of treated hypothyroidism on HRV parameters, without HRV parameters in the time or frequency domain, without a control group, on animals, on children, conferences, congresses, or seminars. Studies had to be primary research. We manually searched the reference lists of all publications with our inclusion criteria to identify studies that would not have been found in the electronic search. We also performed searches within references of included articles or review found using our search strategy, to identify other potentially eligible primary studies. Our search strategy is shown in Fig 1 and S1 Fig. Two authors (VB and RB) conducted the literature searches, reviewed the abstracts and articles independently, checked suitability for inclusion, and extracted the data. When necessary, disagreements were solved with a third author (FD).

Fig 1

Flow chart.

Data extraction

The primary endpoint was the analysis of HRV parameters in untreated hypothyroid patients and in healthy controls. Linear methods are the most traditional measurement of HRV, including time and frequency domains [3]. In the time domain, the RR intervals (or normal-to-normal intervals-NN), the standard deviation of RR intervals (SDNN), the root mean square of successive RR-intervals differences (RMSSD) and the percentage of adjacent NN intervals varying by more than 50 milliseconds (pNN50) were analysed. The frequency domain can be separated in three components according to its frequency ranges [3]: low frequency (LF, 0.04 to 0.15 Hz), high frequency (HF, 0.15 to 0.4 Hz), and very low frequency (VLF, 0.003 to 0.04 Hz). Power is the energy found in a frequency band [21]. LF, HF, and VLF bands are obtained either with the fast Fourier transform algorithm or with autoregressive modelling [3]. LF and HF powers are absolute powers, reported in units of ms2 (square milliseconds). LFnu and HFnu are relative power, called normalized power, in the LF and HF bands, a derived index that is calculated by dividing LF or HF by an appropriate denominator representing the relevant total power: LFnu = LF / (LF + HF) and HFnu = HF / (LF + HF). Due to high inter-individual variability in total and specific band power, LFnu and HFnu allow comparison of frequency domain HRV parameters between two patients [22]. RMSSD and pNN50 are associated with HF and HFnu power, which represents parasympathetic activity, whereas SDNN is associated with LF power, which represents both sympathetic and parasympathetic activity [23]. LFnu emphasizes the control and balance of cardiac sympathetic behaviour [24]. VLF power is also correlated with SDNN measurement due to still uncertain physiological mechanisms [25], thus both sympathetic and parasympathetic activity contribute to VLF power [26, 27]. Total power (TP) and LF/HF ratio, which represented sympathovagal balance, were calculated and reported in this meta-analysis. Secondary outcomes included clinical (BMI, blood pressure, treatments, other diseases), electrical (heart rate), hypothyroidism (duration, etiology, thyroid-stimulating hormone–TSH, free thyroxine–fT4, free triiodothyronine–fT3) and sociodemographic (age, sex, smoking) characteristics (Table 1).

Table 1

Descriptive characteristics of HRV parameters.

HRV parameters
Acronym (unit)	Full name	Signification
Time-domain
RR (ms)	RR–intervals (or Normal to Normal intervals–NN) i.e. beat-by-beat variations of heart rate	Overall autonomic activity
SDNN (ms)	Standard deviation of RR intervals	Correlated with LF power
RMSSD (ms)	Root mean square of successive RR-intervals differences	Associated with HF power and hence parasympathetic activity
pNN50 (%)	Percentage of adjacent NN intervals varying by more than 50 milliseconds	Associated with HF power and hence parasympathetic activity
Frequency-domain
TP (ms2)	Total power i.e. power of all spectral bands	Overall autonomic activity
VLF (ms2)	Very Low Frequency (0.003 to 0.04 Hz)	Thermoregulation, renin-angiotensin system
LF (ms2)	Power of the high-frequency band (0.04–0.15 Hz)	Index of both sympathetic and parasympathetic activity, with a predominance of sympathetic
HF (ms2)	Power of the high-frequency band (0.15–0.4 Hz)	Represents the most efferent vagal (parasympathetic) activity to the sinus node
LF/HF	LF/HF ratio	Sympathovagal balance

Quality of assessment

We used the Scottish Intercollegiate Guidelines Network (SIGN) score, based on different evaluation grids depending on the type of study. For cohort and cross-sectional studies, the evaluation grids are composed of two sections with 4 possible answers (yes, no, can’t say or not applicable): one on the design of the study (14 items) and the other on the overall evaluation of the article (3 items) (S2 Fig) [28]. The “STrengthening the Reporting of OBservational studies in Epidemiology” (STROBE) score is used to check the quality of reports from cohort and cross-sectional studies [29]. By assigning one point per item and subitem, we were able to calculate a percentage of a maximum score of 32 points.

Statistical considerations

We used Stata software (v16, StataCorp, College Station, US) for the statistical analysis [30-34]. Main characteristics were synthetized for each study population and reported as mean ± standard deviation (SD) for continuous variables and number (%) for categorical variables. When data could be pooled, we conducted random effects meta-analyses (DerSimonian and Laird approach) for each HRV parameter comparing patients with untreated hypothyroidism with healthy controls [35]. A negative effect size (ES, standardised mean differences—SMD) [36] denoted lower HRV in patients than in controls. An ES is a unitless measure, centred at zero if the HRV parameter did not differ between hypothyroidism patients and controls. An ES of -0.8 reflects a large effect i.e. a large HRV decrease in patients compared to controls, -0.5 a moderate effect, and -0.2 a small effect. Then, meta-analyses stratified on TSH levels (above and below 10mIU/L or undefined if the TSH level was missing) were performed. We evaluated heterogeneity in the study results by examining forest plots, confidence intervals (CI) and I-squared (I2). I2 is the most common metric to measure heterogeneity between studies, ranging from 0 to 100%. Heterogeneity is considered low for I2<25%, modest for 2550%. We also searched for potential publication bias by examining funnel plots of these meta-analyses. We verified the strength of our results by conducting further meta-analyses after exclusion of studies that were not evenly distributed around the base of the funnel. If the sample size was sufficient, meta-regressions were performed to investigate the relationship between each HRV parameter and relevant clinicobiological parameters (age, sex, blood pressure, BMI, TSH, fT4 levels, fT3 levels). Results were expressed as regression coefficients and 95% confidence intervals (95%CI). P-values less than 0.05 were considered statistically significant.

Results

An initial search produced a possible 863 articles (Fig 1). The number of articles reporting the evaluation of HRV in untreated hypothyroidism was reduced to 17 after elimination of duplicates and use of the selection criteria [15–17, 37–48]. All included articles were written in English.

Among the 17 studies included, six studies were prospective [16, 17, 40–42, 44], nine were cross-sectional [15, 37, 38, 43, 45–50] and one was retrospective [39]. Included studies were published from 2000 to 2018 and conducted across 3 continents (Asia– 8 studies, Europe– 7 studies, America– 2 studies). All included articles compared HRV parameters of patients with untreated hypothyroidism and healthy controls [15–17, 37–48].

Sample size ranged from 14 [16] to 9134 [39], for a total of 11438 patients: 1163 with untreated hypothyroidism and 10275 healthy controls.

Thyroid function was described clinically and biologically in all studies. TSH levels was reported in all studies except two [43, 50]. Nine articles included hypothyroid patients with TSH >10mIU/L [15–17, 37, 38, 42, 44, 47, 48], five with TSH <10mIU/L [39–41, 45, 49], and one with both [46]. Most studies included newly diagnosed and untreated hypothyroid patients before initiation of therapy [16, 17, 37, 38, 43, 47, 48].

HRV recording was ambulatory, spontaneous breathing with normal daily activity in all studies. Most studies used ECG in the supine position at rest to determine HRV [15, 16, 37–39, 43–45, 47, 48, 50], ranging from 4 [37] to 15 minutes [44], except six studies that used a 24-hour holter-ECG [17, 40–42, 46, 49]. Parameters reported were both time and frequency domains in most studies, except two studies that reported only time domain [40, 49] and one only frequency domain [15].

More details on study characteristics (Table 2), aims and quality of articles, inclusion and exclusion criteria, characteristics of population, characteristics of hypothyroidism, and HRV measurements and analysis are described in S3 Fig.

Table 2

Characteristics of included studies.

Study	Country	Design	Subgroup	Untreated hypothyroidism						Healthy controls			ECG, min	HRV parameters
				n	Age, years	Sex, %men	FT4, pmol/L	FT3, pmol/L	TSH, mIU/L	n	Age, years	Sex, %men
Ahmed 2010	Bangladesh	Cross-sectional	Overt	30	38.0 ± 1.2	0.0%	5.1 ± 1.9	-	38.2 ± 30.5	30	36.0 ± 2.6	0.0%	5	TP, LF, HF, LF/HF
Cacciatori 2000	Italy	Prospective	Lying–overt	7	52.1 ± 5.3	0.0%	3.1 ± 0.4	-	55.5 ± 3.5	7	52.0 ± 5.2	0.0%	10	RR, TP, LF, HF, LF/HF
			Standing–overt
Celik 2011	Turkey	Prospective	Subclinical	40	48.0 ± 13.0	10.0%	11.6 ± 3.9	4.0 ± 1.1	6.2 ± 1.2	31	51.0 ± 12.0	9.7%	1440	RR, SDNN, RMSSD
Falcone 2014	Italy	Cross-sectional	Subclinical	55	71.0 ± 13.1	23.6%	24.5 ± 9.0	4.0 ± 1.2	5.4 ± 1.4	170	71.0 ± 12.4	34.7%	1440	RR, SDNN, RMSSD, pNN50
Galetta 2006	Italy	Prospective	Subclinical	42	53.2 ± 14.2	0.0%	9.3 ± 1.1	4.3 ± 0.2	9.8 ± 1.7	30	51.4 ± 16.2	30.0%	1440	RR, SDNN, RMSSD, pNN50, LF, HF, LF/HF
Galetta 2008	Italy	Prospective	Overt	31	53.6 ± 11.8	29.0%	0.7 ± 0.1	1.8 ± 0.3	56.2 ± 14.7	31	50.4 ± 15.3	29.0%	1440	RR, SDNN, RMSSD, pNN50, LF, HF, LF/HF
Gupta 2017	Nepal	Cross-sectional	Subclinical	30	32.0 ± 9.1	33.3%	-	-	22.8 ± 3.5	30	29.3 ± 6.2	33.3%	5	SDNN, RMSSD, pNN50, TP, LF, HF
Heemstra 2010	The Netherlands	Prospective	Overt	11	45.5 ± 10.0	36.4%	1.4 ± 0.7	0.1 ± 0.2	142.4 ± 34.4	21	45.5 ± 8.7	38.1%	15	RR, LF, HF, VLF, LF/HF
Hoshi 2018	Brazil	Cross-sectional	Subclinical	44	55.0 ± 4.0	40.9%	14.2 ± 1.3	4.9 ± 0.4	4.8 ± 1.0	509	52.0 ± 6.5	56.6%	10	SDNN, RMSSD, pNN50, LF, HF, LF/HF
			Overt	59	-	-	-	-	8.7 ± 3.2
Karthik 2009	India	Cross-sectional	Overt	15	29.2 ± 5.7	0.0%	4.0 ± 1.7	2.2 ± 0.8	88.5 ± 20.3	15	27.8 ± 6.6	0.0%	4	RR, SDNN, RMSSD, TP, LF, HF, LF/HF
Mavai 2018	India	Cross-sectional	Overt	35	37.3 ± 9.3	-	9.0 ± 3.7	2.6 ± 1.0	16.9 ± 7.4	25	34.5 ± 10.1	-	5	SDNN, RMSSD, pNN50, TP, LF, HF
Moldabek 2011	Kazakhstan	Cross-sectional	Overt	42	-	-	-	-	32.0 ± 10.2	30	-	-	5	RR, SDNN, RMSSD, pNN50, LF/HF
Peixoto de Miranda 2018	Brazil	Retrospective	Subclinical	511	52.0 ± 6.5	47.2%	-	-	5.1 ± 1.0	8623	50.0 ± 6.0	48.4%	10	RR, SDNN, RMSSD, pNN50, LF, HF
Sahin 2005	Turkey	Cross-sectional	Subclinical (TSH 4.4–9.9mIU/L)	18	41.1 ± 12.6	11.1%	-	-	7.2 ± 3.9	28	41.1 ± 15.2	7.1%	1440	SDNN, RMSSD, pNN50, LF, HF, LF/HF
			Subclinical (TSH>10mIU/L)	13	41.1 ± 12.6	7.7%	-	-	20.6 ± 9.1
Syamsunder 2013	India	Cross-sectional	Overt	54	27.2 ± 4.7	0.0%	8.0 ± 3.6	2.3 ± 0.8	97.6 ± 55.8	50	25.5 ± 5.6	0.0%	10	RR, SDNN, RMSSD, pNN50, TP, LF, HF, LF/HF
Syamsunder 2016	India	Cross-sectional	Subclinical	81	27.3 ± 3.2	0.0%	15.4 ± 6.6	4.1 ± 1.3	12.7 ± 2.3	80	36.6 ± 4.8	0.0%	10	RR, SDNN, RMSSD, pNN50, TP, LF, HF, LF/HF
Xing 2001	China	Prospective	Overt	38	51.0 ± 13.0	23.7%	0.2 ± 0.1	0.9 ± 0.1	65.0 ± 25.6	21	52.0 ± 11.0	23.8%	1440	SDNN, RMSSD, pNN50, LF, HF, LF/HF

FT4: free thyroxine, FT3: free triiodothyronine, TSH: thyroid-stimulating hormone, RR: RR intervals (or normal-to-normal intervals-NNs), SDNN: standard deviation of RR intervals, pNN50: percentage of adjacent NN intervals differing by more than 50 milliseconds, RMSSD: the square root of the mean squared difference of successive RR-intervals, TP: total power, LF: low frequency, HF: high frequency, VLF: very low frequency, LF/HF ratio: low frequency / high frequency ratio.

Meta–analyses of HRV values in untreated hypothyroidism

The main results of the meta-analysis are shown in Fig 2. In comparison to healthy controls, we noted strong evidence (p <0.001) that hypothyroid patient had significantly lower SDNN (ES = -1.27, 95% CI -1.72 to -0.83), RMSSD (-1.66, -2.32 to -1.00), pNN50 (-1.41, -1.98 to -0.84), TP (-1.55, -2.1 to -1.00), LF power (-0.58, -0.89 to -0.28), HF power (-0.98, -1.44 to -0.51), HFnu (-1.21, -1.78 to -0.63) and higher LFnu (1.14, 0.63 to 1.66) and LF/HF ratio (1.26, 0.71 to 1.81). There was no significant difference in RR intervals between hypothyroid patients and healthy controls (p = 0.174) (S4 Fig).

Fig 2

Meta-analysis of heart rate variability parameters of untreated hypothyroid patients compared with controls.

Meta-analysis stratified by TSH levels

RR intervals and LF/HF were only altered in the most severe patients (TSH >10mIU/L) (ES = 0.53, 95% CI 0.09 to 0.96 and 1.34, 0.69 to 2.00, respectively), and not when TSH levels were <10mIU/L (-0.72, -1.52 to 0.07 and 0.56, -0.29 to 1.41, respectively). Despite non-significant comparisons between subgroups, we noted a global higher decrease in HRV when TSH was >10mIU/L: SDNN (-1.17, -1.63 to -0.70 for TSH>10mIU/L and -0.77, -1.23 to -0.31 for TSH<10mIU/L subgroup), RMSSD (-1.13, -1.84 to -0.43 and -1.49, -2.49 to -0.48), pNN50 (-1.19, -1.75 to -0.64 and -0.73, -1.43 to -0.03), LF power (-0.97, -1.68 to -0.25 and -0.35, -0.66 to -0.05) and HF power (-1.02, -1.8 to -0.26 and -0.96, -1.68 to -0.25) (p <0.05). Other parameters were only measured in the most severe patients (TSH >10mIU/L), precluding comparisons between the two subgroups based on TSH levels. However, they were strongly altered (ES greater than 0.80 or -0.80) in those severe patients: TP (-1.70, -2.32 to -1.07), and HFnu (-1.37, -2.01 to -0.73) and higher LFnu (1.28, 0.73 to 1.83) (S4 Fig). All meta-analyses had a high degree of heterogeneity (I2>50%).

Meta–regressions and sensitivity analyses

An increase in fT3 was associated with lower RR intervals (coefficient = -0.75, 95%CI -1.44 to -0.07) (p <0.05). Age was associated with lower RMSSD (-0.09, -0.17 to -0.004) (p = 0.041). Men had lower LFnu (-4.36, -8.53 to -0.19, per % men) and LF/HF (-6.08, -9.52 to -2.64) (p <0.05). An increase in systolic blood pressure was associated with lower HFnu (-0.08, -0.15 to -0.01) and an increase in diastolic blood pressure was associated with lower LF power (-0.25, -0.45 to -0.06) (p <0.05). No significant results were observed for BMI, fT4 and TSH levels (Fig 3 and S5 Fig).

Fig 3

Meta-regressions of significant factors influencing heart rate variability in untreated hypothyroid patients (exhaustive metaregressions are presented in S5 Fig).

The meta-analyses were rerun after excluding studies that were not evenly distributed around the base of the funnel (S6 Fig) and showed similar results.

Discussion

The main results showed a decreased HRV in patients with hypothyroidism that may be explained by the deleterious effect of TSH. The increase in sympathetic and decrease in parasympathetic activity may have clinical implications. Some other factors, such as age or BMI, should also be considered from a clinical perspective.

Deleterious effects of hypothyroidism on HRV

Hypothyroidism is often considered to influence the autonomic nervous system in the opposite direction to hyperthyroidism [51]. Based on clinical data, a decrease in sympathetic activity would be suggested [16]. However, production, release and plasma degradation of catecholamines is increased in hypothyroidism, explaining increased sympathetic activity [14, 52]. These data suggest desensitisation of catecholamine receptors or post-receptor sites in hypothyroidism [16, 53, 54], with reduced binding of β- and α2-adrenergic receptors in cardiac myocytes [53, 54]. These results are consistent with the increased muscle sympathetic activity in hypothyroidism [55]. Similarly, the decreased parasympathetic activity in hypothyroidism may be explained by neuroterminal alteration of cardiac parasympathetic neurons and thus, a decrease in muscarinic effect [56, 57]. Vagal inhibition is more intense than increased sympathetic activity, with a greater decrease in HF power than LF power. Logically, TP decreases markedly (cardiac vagal control) as HF is its main contributor–two third, while LF and VLF contributes one third [3, 58]. HRV is decreased mainly because of a large decrease in vagal activity [3, 58]. No differences in RR intervals is common in hypothyroidism [59], this is in line with our results. The hypothalamus is involved in cardiac autonomic control and TSH release [60, 61], linking the thyroid to the autonomic nervous system [62, 63]. In hypothyroidism, the cardiac autonomic alteration may take place at an hypothalamic level [64]. Indeed, some studies suggested that TSH stimulates sympathetic output from the central nervous system and acts as a neurotransmitter, playing a critical role in determining sympathovagal imbalance [65]. It corroborates the greater HRV decrease in patients with higher TSH levels [45, 46].

Clinical implications

Decreased vagal tone and increased sympathetic activity in hypothyroidism have important clinical implications. Catecholamine receptor desensitization results in a decrease cardiac output, leading to a compensatory increase in norepinephrine release [66]. Hypothyroidism is associated with an increased risk of cardiovascular mortality [67], coronary artery disease [49], and potentially fatal arrhythmias [68, 69]. These complications result from multiple mechanisms (reduced systolic function, diastolic hypertension, atherogenic profile), but also sympathovagal imbalance [41, 69]. Indeed, patients with low vagal tone are more susceptible to cardiovascular diseases such as myocardial infarction, rhythm disorders, and hypertension [70, 71]. It has also been shown that decreased TP predicts an increased risk of sudden cardiac death [72] and total cardiac mortality [73], and that decreased LF was a strong predictor of sudden death independently of other variables [74]. These data suggest that HRV parameters may be a marker of increased mortality in hypothyroid patients [40]. The cardiac effects of hypothyroidism depend on the severity of the disease [65], with higher TSH levels associated with a higher risk of sudden cardiac death [75]. Therefore, it may be worthwhile to consider treatment of hypothyroidism, even for TSH <10mIU/L. However, reversibility of HRV abnormalities in hypothyroidism is not yet demonstrated to prevent cardiac complications.

Other variables related to HRV in hypothyroidism

An increase in fT3 was associated with lower RR, which seems logical as thyroid hormones increase intrinsic activity of the sinus node and thus heart rate [76]. Men were associated with lower LF/HF ratio. This may be explained by the fact that men have lower sympathetic activity and higher parasympathetic activity compared to women [77], hence a decrease in LF/HF ratio [78, 79]. The sympathovagal imbalance could be due to a change in lipid profile as dyslipidemia is common in hypothyroidism [6], and is associated with increased sympathetic activity [80, 81]. However, this variable could not be explored in our meta-analysis due to lack of data. Age was associated with a decreased RMSSD. Indeed, the levels of the HRV time domain parameters decrease with age, especially after 50 years [82, 83] and the prevalence of hypothyroidism increases with age up to 10–15% in elderly patients [4]. We demonstrated that increased diastolic and systolic blood pressure were associated with decreased LF and HFnu power, respectively. The disturbance in blood pressure balance in hypothyroidism with systolic hypotension and diastolic hypertension, possibly reflects an alteration of the autonomic nervous system [84].

Limitations

All meta-analyses have limitations, including those of the individual studies that comprise them, and are theoretically subjected to publication bias [85]. Although the meta-analysis was based on a moderate number of studies [86], the use of broader keywords in the search strategy limits the number of missing studies. The included studies were of variable quality despite our inclusion criteria [39, 50]. Most studies were cross-sectional [15, 37, 38, 43, 45–50], precluding robust conclusions for our meta-analyses [86]. Data collection, inclusion criteria and exclusion criteria were not identical in each study, although similar, which may have affected our results [87]. We limited the influence of extreme results and heterogeneity by exclusion of outliers [88, 89]. In addition, all studies except one [39] were monocentric, limiting the generalizability of our results [87]. Moreover, declarative data from studies are a putative bias [85]. Studies also differed in measurement conditions, such as in duration of recording of HRV parameters [38, 46]. No included studies assessed pulse-based HRV that seems to be less accurate than ECG-based HRV [90]. The interpretation of the LF/HF ratio is controversial in the literature, and may not correspond exactly to the sympatho-vagal balance [91, 92]. Ideally, the sympatho-vagal system tends more towards a non-linear relationship [91, 93]. We did not compute meta-analysis on non-linear assessment of HRV as it has been poorly studied in hypothyroidism. Parasympathetic-sympathetic interactions are complex, non-linear and often non-reciprocal [21]. Thus, non-linear measurements of HRV allow the unpredictability of a time series to be quantified [92], which results from the complexity of the HRV regulatory mechanisms [94-96]. Similarly, VLF power has been investigated by only one study [44] and is recognized as an independent predictor of mortality in patients with heart failure or in chronic hemodialysis patients [97]. The potential importance of VLF in hypothyroidism should be further investigated. Despite most included articles did not show HRV alteration depending on levels of TSH, we showed significant dose response relationship. It may be explained by the fact that each included article only retrieved a small increase in TSH levels, which may explain the absence of significant relation, whereas the combination of all articles in our meta-analysis permitted to analyze a wide range of TSH levels and HRV values. Etiology, duration of hypothyroidism and lipid profile were poorly reported, precluding further analysis. Similarly, the lack of data on spectral analysis of hypothyroidism with TSH below 10mIU/L did not allow conclusion on the type and degree of sympathovagal imbalance.

Conclusion

HRV is markedly decreased in hypothyroid patients. Increased sympathetic and decreased parasympathetic activity may be explained by molecular mechanisms involving catecholamines and by the effect of TSH on HRV parameters. The increased sympathetic and decreased parasympathetic activity may have clinical implications.

(DOC)

Click here for additional data file.

Details for the search strategy used within each database.

(TIFF)

Click here for additional data file.

Quality of included studies.

Methodological quality of included studies using the SIGN checklist. Methodological quality of included studies using the SIGN checklist, by study. SIGN checklist for cohort studies. Methodological quality of included studies using STROBE checklist, by study.

(TIFF)

Click here for additional data file.

Aims of included articles, quality of articles, inclusion and exclusion criteria of included studies, characteristics of population, characteristics of hypothyroidism, and HRV measurements and analysis.

(TIFF)

Click here for additional data file.

Detailed meta–analyses in untreated hypothyroid patients compared with controls for each HRV parameters: RR intervals, SDNN, RMSSD, pNN50, TP, LF, HF, LF/HF.

(TIFF)

Click here for additional data file.

Detailed meta-regressions of factors influencing HRV parameters.

(TIFF)

Click here for additional data file.

Meta funnels.

(TIFF)

Click here for additional data file.

1 Apr 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Daniel M. Johnson, PhD

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating in your Funding Statement:

(This work was supported by the National Natural Science Foundation of China (Grant No. 51475317) and the Shanxi Provincial Natural Science Foundation of China (Grant No.201901D111237).)

Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now.  Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement.

Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

3. Thank you for stating the following in the Acknowledgments Section of your manuscript:

(This work was supported by the National Natural Science Foundation of China (Grant No. 51475317) and the Shanxi Provincial Natural Science Foundation of China (Grant No.201901D111237).)

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

(This work was supported by the National Natural Science Foundation of China (Grant No. 51475317) and the Shanxi Provincial Natural Science Foundation of China (Grant No.201901D111237).)

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

4. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

5. Please amend the manuscript submission data (via Edit Submission) to author Bingbing Peng, and Pengcheng Liu.

6. Please amend your authorship list in your manuscript file to author Bingbing Peng, and Pengcheng Liu.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is an interesting systematic review and meta-analysis heart rate variability (HRV) in hypothyroid patients.

The conclusions from your review that HRV increases with severity of hypothyroidism are not particularly surprising. Much of the data from patients in the studies to which you refer were identified by routine blood test and were subclinical, their ‘hypothyroid status’ the result of deviation from the ‘normal values’.

Peixoto de Miranda et al., (2018) showed that subclinical hyperthyroidism leads to lower heart rate variability. These authors also found that no significant difference subclinical hypothyroidism group when compared to the euthyroid group. Peixoto de Miranda et al., (2018) therefore concluded that the subclinical thyroid dysfunctions presented no relationship with HRV variables. The conclusions from your paper seem to contradict this. Peixoto de Miranda et al., also adjusted for sociodemographic and clinical characteristics in their patients, and this is not apparent from the other references you have used in your review. Perhaps the negative findings from the Peixoto de Miranda et al. study are because the authors had adjusted for sociodemographic and clinical characteristics in their patients. It is well known that thyroid hormones and stress are linked in acute conditions.

In [line 384] you say that "Men were associated with lower LF/HF ratio. This may be explained by the fact that men have lower sympathetic activity and higher parasympathetic activity compared to women". My first question when reading this was the expectation that there would therefore be a sex difference in HRV and hypothyroidism. It is disappointing to see that there is insufficient data in your review to investigate this further.

You don't seem to query the standard interpretation of the LF/HF ratio as sympathetic/ parasympathetic. It is not unequivocally accepted by everyone. (Billman, GE., (2013) The LF/HF ratio does not accurately measure cardiac sympatho-vagal balance. Front. Physiol., 20 February 2013 | https://doi.org/10.3389/fphys.2013.00026).

[Line 425] Your conclusion is seriously deficient. Readers of this journal will be none the wiser by your statement about ‘deleterious effect of TSH on HRV parameters’. This is meaningless. There are a huge number of conditions that change HRV parameters that are not deleterious. Deleterious implies harm.

[Line 426] The statement ‘benefits of HRV assessment in the evaluation and monitoring of the severity of hypothyroidism should be further investigated’ is questionable – to what purpose? HRV can be affected by age, sex, stress, sociodemographic and clinical characteristics, and you have not given a breakdown due to lack of data.

I am happy with most of your interpretation of data. It is a shame that the lack of data precludes an adequate conclusion. It would help the reader if you were to include a list of abbreviations used in the paper. The world of HRV is difficult to understand by the novice.

Reviewer #2: The idea of doing a meta-analysis of HRV and hypothyroidism that includes both short term resting HRV and 24-hour recordings cannot be justified. SDNN from a 24-hour recording and SDNN from a 5-min recording cannot be compared and some of the short recordings even included paced breathing. SDNN is a powerful snapshot of the health of the ANS but only at the 24-hour level where it reflects circadian rhythm and has a lot more to do with sleep quality and the ability of the system to relax during sleep. 5-min or 15-min SDNN has nothing to do with this! More than that, the use of the LF/HF ratio and the concept of sympathovagal balance has been discredited. It is useful only to reflect sudden arousals. LF and HFnu have more validity. Also the report of no heart rate differences between groups makes no sense at all when comparing short term and 24-hour-derived values.

HRV is based on NN intervals and I question the assumption that RR means the same thing.

The authors do not have a deep understanding of HRV and the references seem to be selected without making sure they are current and meaningful. Here is an example:

"High HRV suggests dominant parasympathetic activity [12] with a good ability to adapt and respond to internal and external stimuli [3, 13]."

Ref 12 is from 1993, ref 3 is from 1996 and ref 13 is about autism. Actually high HRV from any recording but especially 24 hrs generally needs to be examined closely. Patients with atrial fibrillation have very high HRV and it is not a good sign. Patients can persent with an erratic sinus rhythm, where the ECG looks fine, and the irregularity associated with the rhythm can appear to be good vagal control of heart rate, but it is not.

Another example of references that are old or appear to have been thrown in without much thought:

Low HRV is an independent predictor of cardiac morbidity [11], due to dominant sympathetic activity [12].

Ref 11 is a comparison of hyperthyroid and euthyroid patients and unlikely to be a primary source for that statement and ref 12, again is from 1993 and a lot has changed. Again no distinction between 24-hr and very short term HRV.

Another quote:

Like a prism refracting light in its different wavelength components, the time domain can be separated in three components according to its frequency ranges [3]: low frequency (LF, 0.04 ± 0.15 Hz), high frequency (HF, 0.15 ± 0.4 Hz), and very low frequency (VLF, 0.003 ± 0.04 Hz).

HRV in the time domain is a purely statistical calculation and the assertion that the time domain can be separated into components is simply wrong. Moreover, for the purposes of this paper, VLF for the short recordings, is not likely to mean anything.

It is truly unfortunate that there is little or no literature about HRV in the non-linear domain in this field. Non-linear HRV tells us about the structure of the HR time series and allows us to distinquish between HRV that is normally organized and HRV that has an excess of disorganized patterns. HRV might be the same in the time domain and without actually examining the plots of the HRV power spectrum, the components, which are simply the area under the curve of power vs. frequency divided into bands, it is impossible to know if HRV is normal.

The review lumps HRV from digitized ECGs and HRV from devices that measure it from a pulse wave based device. Since pulse-based HRV does not have a clear R-wave for beat detection, they are not equivalent. Also, ectopic beats in an ECG-based analysis should be interpolated between the surrounding normal sinus beats, not excluded as stated in the paper.

I could go on, but let me get to the results, which are a noble attempt to summarize the findings but otherwise become boring and unreadable after a short time. There has to be a better way, but when apples are being treated as equivalent to oranges, in terms of how HRV was measured, it gets even worse. The English is good, but saying something to the effect that all studies found something and then after that saying, except two, does not work. It would work better if the authors said all but two studies, because then the reader does not feel like they are being told something "all studies" and then it is taken back.

I have a sense that the authors are highly skilled in actually doing a meta analysis, which is impressive, but really knowing the field and knowing how to cite references in a meaningful way is equally important and that is not here.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Phyllis K Stein

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

28 Apr 2022

Academic Editor

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

[REPLY] Thank you for your comment. We have addressed the comments of the reviewers in a revised manuscript and enclose a point-by-point response.

Please pay attention to all the reviewer's comments, especially those regarding potential bias and references and the use of digitized ECGs and HRVs from devices measuring it from a pulse wave based device. Use of the LF/HF ratio and current discussions regarding the utility of this should also be addressed.

[REPLY] Thank you for your comment. We have tried to respond to the reviewers' comments as best we can. Some references have been changed to more recent ones. All included articles were based on ECG measurements (Holter-ECG or ECG). No articles measuring pulse-based HRV were found in hypothyroidism. The limitations section now reads: "No included studies evaluated pulse-based HRV, which appears to be less accurate than ECG-based HRV [90]." Regarding the use of the LF/HF ratio, we have qualified its use in the limitations session. The limitations section now reads as follows: "The interpretation of the LF/HF ratio is controversial in the literature, and may not accurately reflect sympatho-vagal balance [91,92]. Ideally, the sympatho-vagal system tends more toward a nonlinear relationship [93,94]. We did not calculate a meta-analysis on the nonlinear assessment of HRV because it has been little studied in hypothyroidism. Parasympathetic-sympathetic interactions are complex, nonlinear, and often nonreciprocal [21]. Thus, nonlinear measures of HRV allow quantification of the unpredictability of a time series [92], which results from the complexity of HRV regulatory mechanisms [95-97]."

Journal Requirements

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_ authors_affiliations.pdf

[REPLY] We will adhere to the style requirements according to the style templates sent.

2. Thank you for stating in your Funding Statement:

(This work was supported by the National Natural Science Foundation of China (Grant No. 51475317) and the Shanxi Provincial Natural Science Foundation of China (Grant No.201901D111237).)

Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now.  Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement. Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

[REPLY] The realization of this article has not received any funding. Our article is a meta-analysis, made from a systematic review of the literature. There should be a misunderstanding, or a wrong copy paste when you cite “(This work was supported by the National Natural Science Foundation of China (Grant No. 51475317) and the Shanxi Provincial Natural Science Foundation of China (Grant No.201901D111237)”. Those fundings does not belong to our article.

3. Thank you for stating the following in the Acknowledgments Section of your manuscript:

(This work was supported by the National Natural Science Foundation of China (Grant No. 51475317) and the Shanxi Provincial Natural Science Foundation of China (Grant No.201901D111237).)

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

(This work was supported by the National Natural Science Foundation of China (Grant No. 51475317) and the Shanxi Provincial Natural Science Foundation of China (Grant No.201901D111237).)

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

[REPLY] We added the sentence: "Funding Statement: No funding has to be declared for the realization of this article". Our article is a meta-analysis, made from a systematic review of the literature, and received no specific funding.

4. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

[REPLY] The minimal data set underlying the results are described in the manuscript (S4 Figure). We were unable to access the data for each patient, but only for each included study.

5. Please amend the manuscript submission data (via Edit Submission) to author Bingbing Peng, and Pengcheng Liu.

[REPLY] Sorry, we do not understand who are Bingbing Peng and Pengcheng Liu. I think you may have made a mismatch or a wrong copy paste with a previous article submitted to Plos One. They are not among the authors of our article, and we do not know them …

6. Please amend your authorship list in your manuscript file to author Bingbing Peng, and Pengcheng Liu.

[REPLY] Bingbing Peng and Pengcheng Liu are not among the authors of our article … We do not know them … There should be a misunderstanding from your side.

Reviewer 1

This is an interesting systematic review and meta-analysis heart rate variability (HRV) in hypothyroid patients.

[REPLY] Thank you for your comment.

The conclusions from your review that HRV increases with severity of hypothyroidism are not particularly surprising. Much of the data from patients in the studies to which you refer were identified by routine blood test and were subclinical, their ‘hypothyroid status’ the result of deviation from the ‘normal values’.

[REPLY] Thank you for your comment. Indeed, we did not show that “HRV increases with severity of hypothyroidism” but we did show that “the alteration of HRV increases with the severity of hypothyroidism”. We have shown that this alteration exists when TSH is below 10mIU/L, as well as in cases of subclinical hypothyroidism. Therefore, we concluded a continuum between TSH levels and HRV alterations.

Peixoto de Miranda et al., (2018) showed that subclinical hyperthyroidism leads to lower heart rate variability. These authors also found that no significant difference subclinical hypothyroidism group when compared to the euthyroid group. Peixoto de Miranda et al., (2018) therefore concluded that the subclinical thyroid dysfunctions presented no relationship with HRV variables. The conclusions from your paper seem to contradict this. Peixoto de Miranda et al., also adjusted for sociodemographic and clinical characteristics in their patients, and this is not apparent from the other references you have used in your review. Perhaps the negative findings from the Peixoto de Miranda et al. study are because the authors had adjusted for sociodemographic and clinical characteristics in their patients. It is well known that thyroid hormones and stress are linked in acute conditions.

[REPLY] Thank you for your comment. Patients included in the study by Peixoto de Miranda et al, (2018) had a small increase in TSH (5.1mIU/L on average), which may explain the lack of significant difference. We added the following sentence within the discussion: “Despite most included articles did not show HRV alteration depending on levels of TSH, we showed significant dose response relationship. It may be explained by the fact that each included article only retrieved a small increase in TSH levels, which may explain the absence of significant relation, whereas the combination of all articles in our meta-analysis permitted to analyze a wide range of TSH levels and HRV values.” We did not perform additional meta-regressions other than age, sex, BMI, blood pressure and thyroid function due to lack of data. On socio-demographic characteristics, only male gender is associated with lower LF/HF, which may be explained by lower sympathetic activity and higher parasympathetic activity compared to females (reference 77).

In [line 384] you say that "Men were associated with lower LF/HF ratio. This may be explained by the fact that men have lower sympathetic activity and higher parasympathetic activity compared to women". My first question when reading this was the expectation that there would therefore be a sex difference in HRV and hypothyroidism. It is disappointing to see that there is insufficient data in your review to investigate this further.

[REPLY] Thank you for your comment. Unfortunately, the data from the included studies did not allow us to investigate this issue further.

You don't seem to query the standard interpretation of the LF/HF ratio as sympathetic/ parasympathetic. It is not unequivocally accepted by everyone. (Billman, GE., (2013) The LF/HF ratio does not accurately measure cardiac sympatho-vagal balance. Front. Physiol., 20 February 2013 | https://doi.org/10.3389/fphys.2013.00026).

[REPLY] Thank you for your comment. The Limitations section now reads: “The interpretation of the LF/HF ratio is controversial in the literature, and may not correspond exactly to the sympatho-vagal balance [91,92]. Ideally, the sympatho-vagal system tends more towards a non-linear relationship [93,94].”.

[Line 425] Your conclusion is seriously deficient. Readers of this journal will be none the wiser by your statement about ‘deleterious effect of TSH on HRV parameters’. This is meaningless. There are a huge number of conditions that change HRV parameters that are not deleterious. Deleterious implies harm.

[Line 426] The statement ‘benefits of HRV assessment in the evaluation and monitoring of the severity of hypothyroidism should be further investigated’ is questionable – to what purpose? HRV can be affected by age, sex, stress, sociodemographic and clinical characteristics, and you have not given a breakdown due to lack of data.

[REPLY] Thank you for your comment. We removed the word deleterious and also reworded the conclusion. The conclusion now reads: “HRV is markedly decreased in hypothyroid patients. Increased sympathetic and decreased parasympathetic activity may be explained by molecular mechanisms involving catecholamines and by the effect of TSH on HRV parameters. The increased sympathetic and decreased parasympathetic activity may have clinical implications.”

I am happy with most of your interpretation of data. It is a shame that the lack of data precludes an adequate conclusion. It would help the reader if you were to include a list of abbreviations used in the paper. The world of HRV is difficult to understand by the novice.

[REPLY] Thank you for your comment. We added a new table 1:

HRV parameters

Acronym (unit)

Full name

Signification

Time-domain

RR (ms)

RR–intervals (or Normal to Normal intervals – NN) i.e. beat-by-beat variations of heart rate

Overall autonomic activity

SDNN (ms)

Standard deviation of RR intervals

Correlated with LF power

RMSSD (ms)

Root mean square of successive RR-intervals differences

Associated with HF power and hence parasympathetic activity

pNN50 (%)

Percentage of adjacent NN intervals varying by more than 50 milliseconds

Associated with HF power and hence parasympathetic activity

Frequency-domain

TP (ms2)

Total power i.e. power of all spectral bands

Overall autonomic activity

VLF (ms2)

Power of the Very Low Frequency (0.003 to 0.04 Hz)

Thermoregulation, renin-angiotensin system

LF (ms2)

Power of the Low-Frequency band (0.04 to 0.15 Hz)

Index of both sympathetic and parasympathetic activity, with a predominance of sympathetic

HF (ms2)

Power of the High-frequency band (0.15 to 0.4 Hz)

Represents the most efferent vagal (parasympathetic) activity to the sinus node

LF/HF

LF/HF ratio

Sympathovagal balance

Reviewer 2

The idea of doing a meta-analysis of HRV and hypothyroidism that includes both short-term resting HRV and 24-hour recordings cannot be justified. SDNN from a 24-hour recording and SDNN from a 5-min recording cannot be compared and some of the short recordings even included paced breathing. SDNN is a powerful snapshot of the health of the ANS but only at the 24-hour level where it reflects circadian rhythm and has a lot more to do with sleep quality and the ability of the system to relax during sleep. 5-min or 15-min SDNN has nothing to do with this! […] Also the report of no heart rate differences between groups makes no sense at all when comparing short term and 24-hour-derived values.

[REPLY] Thank you for your comment. The effect sizes for each included article were calculated by comparing patients with hypothyroidism and controls within the same article (please see the methods section: “we conducted random effects meta-analyses (DerSimonian and Laird approach) for each HRV parameter comparing patients with untreated hypothyroidism with healthy controls [35].”) Therefore, we did not compare 5 minutes measures to 24 hours measures. First, we calculated the effect sizes for each included article. For example, we compared 5 minutes measures in untreated hypothyroidism patients and 5 minutes measures in healthy controls, in order to retrieve the corresponding effect size. On another article, we compared 24 hours measures in untreated hypothyroidism patients and 24 hours measures in healthy controls, in order to retrieve again the corresponding effect size. And so on for each included article. Therefore, even if we agree that differences in measurement time is a limitation, our data are valid as we pooled the effect sizes of each individual article to produce on overall results i.e. answering the question whether HRV is altered in patients with untreated hypothyroidism compared with healthy controls (and differences in measurement time are comprised within each effect sizes). The limitations section reads: “Studies also differed in measurement conditions, such as in duration of recording of HRV parameters [38, 46].”

More than that, the use of the LF/HF ratio and the concept of sympathovagal balance has been discredited. It is useful only to reflect sudden arousals. LF and HFnu have more validity.

[REPLY] Thank you for your comment. We totally agree with the discredit towards LF/HF ratio and the concept of sympathovagal balance. We have chosen to let the LF/HF ratio as it seems important to some readers. We added some references about the ongoing debate on the LF/HF ratio in the Limitations. The Limitations section now reads: “The interpretation of the LF/HF ratio is controversial in the literature, and may not correspond exactly to the sympatho-vagal balance [91,92]. Ideally, the sympatho-vagal system tends more towards a non-linear relationship [93,94].”.

HRV is based on NN intervals and I question the assumption that RR means the same thing.

[REPLY] Thank you for your comment. We added a new table 1 to explain the meaning of each HRV parameters. Yes, RR–intervals are also called Normal to Normal intervals – NN, i.e. beat-by-beat variations of heart rate.

The authors do not have a deep understanding of HRV and the references seem to be selected without making sure they are current and meaningful. Here is an example:

"High HRV suggests dominant parasympathetic activity [12] with a good ability to adapt and respond to internal and external stimuli [3, 13]." Ref 12 is from 1993, ref 3 is from 1996 and ref 13 is about autism.

[REPLY] Thank you for your comment. References were selected based on their quality, the reference from 1996 is the task force, still valid to date, and cited in all articles examining HRV. The reference on “Autism” from 2017 has in fact no link with autism, it is a vulgarization of the explanation of HRV. This article of 2017 was also judged sufficiently good to be posted in the Huffington post, for a mass communication. So the references were carefully chosen for readers to allow them both to have a deep understanding of HRV if they want to go in details, or to have a vulgarization for a quick understanding. Reference from 1993 is the first paper to give guidance on the interpretation of low or high HRV. Indeed, this reference is outdated, so we have removed it. For reference 11 of Ramanathan, this is indeed an error on our part. This reference has been replaced. The sentence now reads as follows: “Low HRV is an independent predictor of cardiac morbidity [11], while high HRV suggests good ability to adapt and respond to internal and external stimuli [3,12].”

Actually high HRV from any recording but especially 24 hours generally needs to be examined closely. Patients with atrial fibrillation have very high HRV and it is not a good sign. Patients can persent with an erratic sinus rhythm, where the ECG looks fine, and the irregularity associated with the rhythm can appear to be good vagal control of heart rate, but it is not.

[REPLY] Thank you for your comment. We totally agree with you. As it is a meta-analysis, we followed inclusion of included articles. Hyperthyroidism can indeed give erratic sinus rhythm and atrial fibrillation, but not in hypothyroidism.

Another example of references that are old or appear to have been thrown in without much thought:

Low HRV is an independent predictor of cardiac morbidity [11], due to dominant sympathetic activity [12].

Ref 11 is a comparison of hyperthyroid and euthyroid patients and unlikely to be a primary source for that statement and ref 12, again is from 1993 and a lot has changed. Again no distinction between 24-hr and very short term HRV.

[REPLY] Thank you for your comment. Reference from 1993 is the first paper to give guidance on the interpretation of low or high HRV. Indeed, this reference is outdated, so we have removed it. For reference 11 of Ramanathan, this is indeed an error on our part. This reference has been replaced. The sentence now reads as follows: “Low HRV is an independent predictor of cardiac morbidity [11], while high HRV suggests good ability to adapt and respond to internal and external stimuli [3,12].”

Another quote: Like a prism refracting light in its different wavelength components, the time domain can be separated in three components according to its frequency ranges [3]: low frequency (LF, 0.04 ± 0.15 Hz), high frequency (HF, 0.15 ± 0.4 Hz), and very low frequency (VLF, 0.003 ± 0.04Hz). HRV in the time domain is a purely statistical calculation and the assertion that the time domain can be separated into components is simply wrong.

[REPLY] Thank you for your comment. Indeed, the time domain is a purely statistical calculation and is composed of RMSSD, SDNN, pNN50... We have also indicated this in the sentence: “In the time domain, the RR intervals (or normal-to-normal intervals-NN), the standard deviation of RR intervals (SDNN), the root mean square of successive RR-intervals differences (RMSSD) and the percentage of adjacent NN intervals varying by more than 50 milliseconds (pNN50) were analysed.”. As for the frequency domain, this is a picture to better imagine what the frequency domain is for HRV novices. But indeed, you are right, this sentence is confusing. We have changed the sentence to read as follows: “The frequency domain can be separated in three components according to its frequency ranges [3]: low frequency (LF, 0.04 ± 0.15 Hz), high frequency (HF, 0.15 ± 0.4 Hz), and very low frequency (VLF, 0.003 ± 0.04Hz).”

Moreover, for the purposes of this paper, VLF for the short recordings, is not likely to mean anything.

[REPLY] Thank you for your comment. Indeed, VLF is not measured in short recordings, and we have not performed a meta-analysis on VLF due to lack of data on 24-hour recordings.

It is truly unfortunate that there is little or no literature about HRV in the non-linear domain in this field. Non-linear HRV tells us about the structure of the HR time series and allows us to distinquish between HRV that is normally organized and HRV that has an excess of disorganized patterns. HRV might be the same in the time domain and without actually examining the plots of the HRV power spectrum, the components, which are simply the area under the curve of power vs. frequency divided into bands, it is impossible to know if HRV is normal.

[REPLY] Thank you for your comment. We totally agree. We added some references in favor of non-linear domain. The limitation section now reads: “We did not compute meta-analysis on non-linear assessment of HRV as it has been poorly studied in hypothyroidism. Parasympathetic-sympathetic interactions are complex, non-linear and often non-reciprocal [21]. Thus, non-linear measurements of HRV allow the unpredictability of a time series to be quantified [92], which results from the complexity of the HRV regulatory mechanisms [95-97].”

The review lumps HRV from digitized ECGs and HRV from devices that measure it from a pulse wave based device. Since pulse-based HRV does not have a clear R-wave for beat detection, they are not equivalent. Also, ectopic beats in an ECG-based analysis should be interpolated between the surrounding normal sinus beats, not excluded as stated in the paper.

[REPLY] Thank you for your comment. We totally agree that pulse-based HRV is less reliable than ECG measures. All articles included were based on ECG-measurements (Holter-ECG or ECG). No articles measuring pulse-based HRV was retrieved in hypothyroidism. The limitation section now reads: “No included studies assessed pulse-based HRV that seems to be less accurate than ECG-based HRV [90].”

I could go on, but let me get to the results, which are a noble attempt to summarize the findings but otherwise become boring and unreadable after a short time. There has to be a better way, but when apples are being treated as equivalent to oranges, in terms of how HRV was measured, it gets even worse. The English is good, but saying something to the effect that all studies found something and then after that saying, except two, does not work. It would work better if the authors said all but two studies, because then the reader does not feel like they are being told something "all studies" and then it is taken back.

[REPLY] Thank you for your comment. First we reworded some sentences, and second we take the opportunity to move the descriptive characteristics of included articles in supplementary materials. We now present within the main manuscript a summary of descriptive characteristics. Readers that still want further details can have access to them in supplementary materials. (S3 Figure)

I have a sense that the authors are highly skilled in actually doing a meta analysis, which is impressive, but really knowing the field and knowing how to cite references in a meaningful way is equally important and that is not here.

[REPLY] Thank you for your comment. We removed some old historical references and we now cite more recent reviews or articles from high impact factor journals. We believe our article was strongly improved following your recommendations – particularly your suggestion to summarize the descriptive characteristics of included articles.

Submitted filename: ResponseToReviewers.docx

Click here for additional data file.

18 May 2022

Heart rate variability in hypothyroid patients: A systematic review and meta-analysis

PONE-D-22-02213R1

Dear Dr. Brusseau,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Daniel M. Johnson, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: I belive this it now much improved, acceptable and worth publishing. However, mea culpa, I did not have the energy to again go thru it line by line to see if there was anything else :-( However, my comments were politely and responsively addressed.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Dr Duncan Banks

Reviewer #2: Yes: Phyllis K. Stein

23 May 2022

PONE-D-22-02213R1

Heart rate variability in hypothyroid patients: A systematic review and meta-analysis

Dear Dr. Brusseau:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Daniel M. Johnson

Academic Editor

PLOS ONE