Short-term deceleration capacity of heart rate predicts post-induction hypotension in patients with low ASA status: a prospective observational study

Abstract

Background

Autonomic dysfunction is a risk factor for hypotension after anesthesia induction. Deceleration capacity of heart rate (DC) is a new method to evaluate autonomic function. This prospective observational study was designed to evaluate whether the deceleration capacity of heart rate measured by a 5-min preoperative Electrocardiogram (ECG) can reliably predict post-induction hypotension (PIH).

Methods

Patients aged 18 to 65 undergoing elective surgery with lower ASA status I or II were included in this study. DC, root mean square density (RMSSD) and low frequency/high frequency ratio (LF/HF) were calculated from 5-min segments of ECG measured in the quiet state before surgery. PIH was defined as mean arterial pressure (MAP) < 65 mmHg or a decrease of > 30% for at least 1 min from induction of anesthesia to 10 min after tracheal intubation. Patients were divided into PIH and non-PIH groups according to whether they developed PIH or not.

Results

A total of 141 patients were enrolled in this study, of whom 63 (44.7%) presented with PIH. The RMSSD (p = 0.036) and DC (p < 0.001) of the PIH group were smaller, and the LF/HF was higher (p = 0.039). After adjusting for confounding factors (Model 2), DC was identified as an independent predictor of PIH (Odds Ratio: 0.377). The receiver operating characteristic (ROC) analysis showed that DC had a good diagnostic value as a predictor (AUC: 0.777; 95%CI: 0.705–0.909; p < 0.001).

Conclusions

These results suggest that DC measured by ECG 5 min before anesthesia can predict PIH to some extent in patients with ASA status I or II.

Trial registration

Chinese Clinical Trial Registry, identifier: ChiCTR2400094595, Date: 25/12/2024.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12871-025-03262-0.

Induction

Intraoperative hypotension (IOH) can lead to a series of adverse outcomes, including acute kidney injury, myocardial injury [1], and ischemic stroke [2]. Uncontrolled IOH will not only prolong the length of hospital stay [3], but also affect the prognosis of patients [4] and even increase the mortality [5]. Post-induction hypotension (PIH) occurs between induction of anesthesia and skin incision. A significant portion of overall intraoperative hypotension (IOH) occurs during the post-induction period. A previous study showed that PIH accounted for 52% of IOH [6]. Therefore, anesthesiologists should try to prevent the occurrence of PIH.

Traditionally, hypovolemia is considered a risk factor for PIH [7]. However, fluid optimization alone cannot effectively prevent PIH [8], which means that PIH is affected by multiple factors. The study by Karim Kouz et al. divided PIH into six endotypes, among which the endotype myocardial depression was the most common endotype [9]. Jian et al. similarly included myocardial depression and bradycardia in the four hypotensive endotypes [10]. Dynamic regulation of cardiac output depends on autonomic function [11]. Unfortunately, the status of preoperative autonomic function is not well studied. Heart rate variability (HRV), such as root mean square density (RMSSD) and low frequency/high frequency ratio (LF/HF), is currently the leading indicator in evaluating autonomic nerve function, but it is susceptible to interference by many factors [12]. The maximum pupillary contraction velocity can reflect the parasympathetic function to a certain extent [13], but it has high requirements for the measurement environment and is not simple and quick enough.

Deceleration capacity refers to the prolongation of the subsequent RR interval (the time interval between two consecutive R waves on the electrocardiogram) compared with the previous RR interval during two consecutive cardiac cycles, and represents the ability of the parasympathetic nervous system to regulate the cardiovascular system. Deceleration capacity calculated by phase-rectified signal averaging (PRSA) can reduce the non-periodic component in complex time series, quantitatively assess changes in vagal tone [14], and is more sensitive and specific in predicting adverse cardiovascular events than traditional measures of left ventricular ejection fraction and HRV [15].

In several recent studies, short-term DC has been confirmed to have a good correlation with perioperative vagal activity [12, 16]. Assuming that DC obtained by a 5-min electrocardiogram (ECG) is related to hemodynamic instability after induction, we aimed to evaluate whether short-term DC could be used as a simple method to predict PIH in patients with low ASA status.

Materials and methods

Patients

This observational study received ethical approval from the Ethics Committee of Jinling Hospital, a tertiary teaching hospital, with approval number 2024DZKY-092–02, and was officially registered in the Chinese Clinical Trial Registry with registration number ChiCTR2400094595 (registration date: 25/12/2024). All participants provided written informed consent.

Inclusion criteria for participants were as follows: age between 18 and 65 years, ASA class I-II, and elective general anesthesia. Participants were excluded if they had any of the following characteristics: arrhythmias (e.g., atrial fibrillation, ventricular premature beat, atrioventricular block, etc.); use medications that may affect HRV; the presence of a disease affecting the autonomic nervous system (e.g., diabetes); hypertension; the presence of heart disease (e.g., valvular heart disease, cardiomyopathy); surgery is planned in the lateral or prone position.

Electrocardiographic measurements

After all participants rested in the supine position for at least 5 min in the anesthesia preparation room, a 10-min supine ECG was obtained using a Micro-ECG recorder (HealinK-R211b, HeaLink Ltd, Bengbu, China) at a rate of 400 Hz. The signal broadband was 0.6-40 Hz. During the measurement, the patient was instructed to remain quiet for 10 min and to breathe naturally without speaking or moving.

We checked for any abnormalities in signal quality and, if any, excluded cases. ECG R-R Beat-to-Beat Interval time series were extracted based on the Pan-Tompkins algorithm and further analyzed using Kubios (version 4.1.2) software. Kubios HRV is an advanced computer program for HRV extraction and analysis developed by the research team of the University of Eastern Finland based on MATLAB software. We selected the segments with no movement or the least movement within 5 min to obtain relatively stable heart rate variability parameters, including RMSSD and frequency domain parameters calculated by fast Fourier transform (FFT). The low frequency (LF) and high frequency (HF) signals were normalized to reduce the interference of very low frequency (VLF).

The calculation method of RMSSD is divided into the following steps: Firstly, obtain the sequence of RR intervals. Secondly, calculate the difference between consecutive RR intervals. Then, square each difference, calculate the average of the squared differences, and finally, take the square root of the average. The specific calculation formula is: .

The calculation method of LF/HF is divided into the following steps: Firstly, obtaining high-quality RR interval sequences; Secondly, using fast Fourier transform to perform spectral estimation on the processed equidistant RR intervals; Thirdly, defining frequency bands and calculating LF and HF power respectively (LF defined as 0.04–0.15 Hz, HF defined as 0.15–0.4 Hz); Finally, calculating the LF/HF ratio.

DC is regarded as a new HRV parameter. It is calculated based on the PRSA algorithm provided by Baver and is specifically divided into the following steps: Firstly, define the interval longer than the previous heartbeat interval as the anchor point to determine the deceleration period. Secondly, select the interval data segments around the anchor point as the heart rate segments (all heart rate segments should have the same size, and the heart rate segments around adjacent anchor points can overlap). Next, arrange and phase the sequence of each heart rate segment to align them at the anchor point. Then, the PRSA sequence number period is averaged to obtain X (0), X (1), X (−1), and X (−2), where X (0) is the average RR interval of all anchor points, X (1) and X (−1) are the average RR intervals of the right and left adjacent anchor points, and X (−2) is the average RR interval of the second left adjacent anchor point. Finally, calculate using the formula DC = [X (0) + X (1)—X (−1)—X (−2)]/4 [15]. Further biophysics calculation equations are in the supplement.

Anesthesia management

All patients began fasting at midnight without pre-medication and received Ringer’s acetate infusion at a rate of 10 mL/kg/h after entering the operating room. Vital signs (electrocardiogram, respiratory rate, blood pressure, pulse, airway pressure, and end-tidal CO2) were monitored. The anesthesia induction protocol followed a standard routine sequence of induction: 0.04 mg/kg midazolam, 0.3ug/kg sufentanil, 2 mg/kg propofol, and 0.15 mg/kg cisatracurium. After 3 min of muscle relaxation, endotracheal intubation was performed by an experienced anesthesiologist using a video laryngoscope.

Blood pressure measurements

Invasive blood pressure measurements were performed after the patient entered the operating room and before the induction of anesthesia began. After local infiltration with lidocaine, invasive blood pressure monitoring was performed by inserting a 22-gauge arterial catheter (Supercath Ztu-V, Japan) at the styloid process where the radial artery pulsated most prominently (1-2 cm). The catheter was then connected to a pressure transducer (Hisern, Zhejiang) and rinsed with heparin water. A monitor recorded MAP every minute (Mindray, China).

Sample size calculations and data collection

The sample size was determined using the following formula: N = [(Z_α/2 + Z_β) S/δ]². The standard deviation was obtained according to the reference [16], and the allowable error was δ = (0.25–0.5) S. In this case, 0.25 was taken, and the required number of cases was calculated to be 126 (α = 0.05, power = 80%). Considering the 10% dropout rate, about 140 patients were planned to be enrolled.

The basic characteristics of the patients were obtained by questionnaire, including age, gender, height, weight, gender, and comorbidities. Plus, MAP and heart rate (HR) were recorded every minute from induction until 10 min after intubation, and baseline MAP was defined as the blood pressure value at 1 min before induction. PIH was defined as MAP < 65 mmHg or a decrease in MAP of more than 30% from baseline for ≥ 1 min in the interval between induction and 10 min after intubation. Patients were divided into hypotension and non-hypotension groups according to whether hypotension occurred.

Statistical analysis

Data collected were recorded using Microsoft Excel (v2304, Microsoft, USA). The Kolmogorov–Smirnov test was used to assess the normality of the collected data. Measurement data with normal distribution were reported as mean ± standard deviation ( ± s) and compared between groups using an independent sample t-test. Non-normal distribution data were expressed as medians (interquartile range), and the Mann–Whitney U test was used to compare the differences. For categorical variables, the analysis was performed using the chi-square test, with results presented as numerical values and percentages.

According to Bauer et al., the DC frequency distribution was tabulated [15], and the significance was analyzed using the simple formula z = (a-b)/√a + b proposed by Pocock [17]. Binary logistic regression analysis was performed to explore the relationship between HRV parameters and the incidence of PIH. Given the close relationship between age and autonomic function, age was included separately as a variable in Model 1 [18]. Age, gender, albumin level, body mass index (BMI), ASA grade, and baseline MAP were included in model 2 for analysis based on the results of univariate analysis, clinical experience, and a retrospective study of other large samples [6, 19, 20]. Then, according to the analysis results, the receiver operating characteristic (ROC) curve analysis was used to detect the predictive ability of the indicators of deceleration capacity of heart rate for hypotension after induction.

All statistical analyses were performed using IBM SPSS (version 26.0 IBM Corporation, USA). p < 0.05 was considered statistically significant.

Results

A total of 155 patients were initially included in the study, of which five patients with hypertension and diabetes were excluded, two patients were excluded due to inability to remain still during measurement, and seven patients dropped out of the trial due to multiple catheter placement failures or loss of measurement data. Ultimately, data from 141 patients were analyzed, of whom 63 (44.7%) developed PIH. Surgical procedures included gynecological (n = 14), general (n = 30), otolaryngology (n = 14), urologic (n = 20), and orthopedic procedures (n = 63) (Fig. 1).

Fig. 1.

Study flow chart

Patient data

Notably, there were no differences between patients with PIH and those without PIH in gender, BMI, ASA status, past medical history, red blood cell count, hemoglobin and albumin levels (p > 0.05). However, patients in the PIH group were older than those in the non-PIH group (p = 0.022) and had a lower preoperative baseline MAP than those in the non-PIH group (Table 1).

Table 1.

Patient baseline characteristics

	PIH (n = 63)	Non-PIH (n = 78)	P Value
Age (years)	49 [36–55]	36 [28–52]	0.022
18–24	8 (12.7)	11 (13.6)
25–34	7 (10.9)	26 (33.3)
35–44	10 (16.4)	15 (19.7)
45–54	21 (32.7)	14 (18.2)
55–64	17 (27.3)	12 (15.2)
Gender (male/female)	33/30	43/35	0.865
Height, cm	168.5 ± 8.0	168.2 ± 8.5	0.832
BMI, kg/m^2	23.7 ± 3.4	23.3 ± 2.9	0.440
ASA (I/II)	6/57	8/70	0.885
History of smoking, n (%)	6 (9.5)	12 (15.4)	0.327
History of drinking, n (%)	10 (15.9)	8 (10.3)	0.320
Aortic sclerosis	10 (15.9)	6 (7.7)	0.128
Red blood cells, 10^12/L	4.4 [4.10–5.1]	4.5 [4.2–5.0]	0.839
Hemoglobin, g/L	135.6 ± 20.6	132.2 ± 19.8	0.369
Hematocrit	0.41 [0.37–0.46]	0.41 [0.38–0.44]	0.405
Albumin, g/L	43.0 ± 5.4	43.1 ± 4.8	0.951

Normally distributed results were reported as mean ± standard deviation ( ± s), while non-normally distributed data were expressed as medians [interquartile ranges]. Categorical data were reported as numbers (%)

BMI body mass index, ASA American Society of Anesthesiologists physical status

Compared with patients without PIH, the DC value (p < 0.001) and RMSSD value (p = 0.036) in the PIH group were lower, and the LF/HF value was higher (p = 0.039) (Table 2). The frequency band distribution of DC values is shown in Table 3. According to the DC grading criteria of Bauer et al. [15], the proportion of intermediate risk population in the PIH group was higher (P < 0.001), suggesting that the vagal tone of the population was weak. However, the proportion of low risk group was higher in the non-PIH group (P < 0.01), indicating that the vagal tone of the population was strong. There was no significant difference in the high-risk population between the two groups (P > 0.2).

Table 2.

Hemodynamic and ECG measurement comparisons

Measure	PIH (n = 63)	Non-PIH (n = 78)	P Value
Baseline MAP, mmHg	90.3 ± 12.5	97.3 ± 11.9	0.002
Baseline HR, beats/min	78 ± 15	76 ± 13	0.415
Decrease in MAP (%)	31.0 ± 7.9	20.5 ± 13.5	< 0.001
Percentage change in HR (%)	5.3 ± 17.8	3.4 ± 11.3	0.602
RMSSD	23.88 ± 23.72	31.84 ± 20.72	0.036
DC	5.42 ± 1.89	7.66 ± 2.45	< 0.001
LFnu	66.33 ± 17.02	62.92 ± 15.50	0.216
HFnu	33.64 ± 17.02	37.02 ± 15.48	0.219
LF/HF	3.09 ± 2.52	2.34 ± 1.78	0.039

Normally distributed results were reported as mean ± standard deviation ( ± s)

MAP mean arterial pressure, HR heart rate, RMSSD root mean square density, DC deceleration capacity, LF/HF low frequency/high frequency ratio, LFnu low frequency normalized units, HFnu high frequency normalized units

Table 3.

The frequency band distribution of DC values

DC	n	PIH (n = 63)	Non-PIH (n = 78)	P
< 2.5	4	3 (75%)	1 (25%)	> 0.2
2.5- < 4.5	23	20 (87%)	3 (13%)	< 0.001
≥ 4.5	114	40 (35%)	74 (65%)	< 0.01
4.5- < 6.5	45	22 (49%)	23 (51%)	> 0.2
≥ 6.5	69	18 (26%)	51 (74%)	< 0.001

Count data were reported as numbers (%)

DC deceleration capacity

Univariate analysis by gender showed that there was no significant difference in hypotension events (p = 0.865), with a mean DC of 6.77 ± 2.49 in men and 6.58 ± 2.54 in women (p = 0.661). The RMSSD of women was lower than that of men (p = 0.026) (Table 4).

Table 4.

Comparison of the demographic characteristics of the men with women using t test

	male	Female	Significance
Baseline MAP	94.3 ± 12.3	93.9 ± 13.0	0.849
RMSSD	29.72 ± 22.89	21.91 ± 13.96	0.026
DC	6.77 ± 2.49	6.58 ± 2.54	0.661
LF/HF	2.89 ± 2.06	2.56 ± 2.37	0.403
Event (n)	33 (43%)	30 (46%)	0.865

Event, Hypotension was defined as MAP < 65 mmHg or a 30% decrease in MAP and lasted for at least one minute; Normally distributed results were reported as mean ± standard deviation ( ± s); Count data were reported as numbers (%)

RMSSD root mean square density, DC deceleration capacity, LF/HF low frequency/high frequency ratio

Regression analysis

Univariate analysis revealed a relationship between PIH and DC, RMSSD, and LF/HF. Given the specificity of age, after including age as a variable in the adjusted model, only DC was associated with PIH (OR 0.610, 95%CI 0.483–0.771; p < 0.01) (Model 1). After further adjustment for age, sex, ASA physical status, albumin level, BMI, and baseline MAP (Model 2), DC (OR 0.555, 95%CI 0.403–0.763; p < 0.01) were identified as significant independent predictors of PIH, while RMSSD (OR 0.988, 95%CI 0.095–1.019; p > 0.05) and LF/HF (OR 1.138, 95%CI 0.871–1.487; p > 0.05) were not associated with PIH (Table 5). DC was negatively correlated with PIH, and each 1 increase in DC would reduce the incidence of hypotension by 44.5%.

Table 5.

Multivariate logistic regression

Predictors	Unadjusted analysis OR [95%]	Adjusted analysis OR [95%]
		Model 1	Model 2
RMSSD	0.982[0.965–0.999]*	1.006[0.984–1.029]	0.988[0.985–1.019]
DC	0.576[0.462–0.718]**	0.610[0.483–0.771]**	0.555[0.403–0.763]**
LF/HF	1.186[1.003–1.402]*	1.096[0.922–1.302]	1.138[0.871–1.487]

Model 1: Adjusted for age; Model 2: Adjusted for age, sex, ASA physical status, albumin, BMI, and baseline MAP

RMSSD root mean square density, DC deceleration capacity, LF/HF low frequency/high frequency ratio

^*p < 0.05, ^**p < 0.01

ROC curve analysis

Based on the results of logistics regression, we used the receiver operating characteristic (ROC) curve to analyze the predictive ability of DC for PIH, and the results showed that DC showed good diagnostic value as a predictor (AUC: 0.777; 95% CI: 0.699–0.855; P < 0.001) (Fig. 2). When the optimal cut-off value of DC was 6.13, the sensitivity and specificity were 79.5% and 68.3% respectively. When the optimal cut-off value of DC was 5.74, the sensitivity and specificity were 85.9% and 61.9%, respectively. We also analyzed the area under the curve (AUC) of RMSSD, LF/HF, age, and Base MAP, which were 0.696, 0.599, 0.621, and 0.666, respectively.

Fig. 2.

Comparison of Receiver Operating Characteristic (ROC) curves of deceleration capacity (DC), root mean square density (RMSSD), low frequency/high frequency ratio (LF/HF), age, and the baseline MAP to predict PIH

Discussion

In this study, we found that measuring DC 5 min before surgery can help identify patients at higher risk for PIH. Compared with traditional HRV indexes and other factors such as age, DC was an independent predictor of PIH with an AUC of 0.777 (0.699–0.855). Preoperative measurement of DC can provide a simpler and faster method for anesthesiologists to prevent PIH.

Patients with poor preoperative autonomic nerve function are more likely to cause hemodynamic instability after induction. Parasympathetic excitation usually causes slowing of heart rate and lowering of blood pressure [13]; DC and RMSSD are closely related to parasympathetic activity [12, 21], LF/HF reflects the balanced state of sympathetic and parasympathetic nerves, and an increased ratio suggests sympathetic predominance [22]. However, in this trial, we found that patients in the PIH group had lower DC and RMSSD, while higher LF/HF, suggesting that patients with good parasympathetic activity are less prone to PIH. This may be because hemodynamic stability in patients depends more on the co-regulation of autonomic nerves. After induction of anesthesia, sympathetic nerve activity is more strongly inhibited due to the action of anesthetics [23], and patients with dominant sympathetic nerve activity and low parasympathetic nerve activity before surgery cannot adjust their autonomic nerve function in time, so they are more prone to extreme hemodynamic fluctuations.

Short-term HRV and its derivatives have been shown to be reliable tools for predicting hypotension after spinal anesthesia [12, 24]. Still, HRV is widely criticized for being susceptible to interference by age, gender, obesity, comorbidities, respiration, and circadian rhythm [12]. A systematic review showed that the same index of short-term HRV could reach a maximum of 118% variation in different trials [25]. In this trial, there was a significant difference between RMSSD and LF/HF in the PIH group and the non-PIH group (p < 0.05). However, RMSSD and LF/HF could not be used as predictors of PIH after excluding the age interference factor. A meta-analysis by Koenig et al. suggested that researchers of heart rate variability should pay attention to the gender difference of patients [26]. In the study by Lihui Zheng et al., men and women had different DC cutoff values [27]. The results of the univariate analysis of gender showed that there was a significant difference in RMSSD between males and females (p < 0.05) but no significant difference in DC (p = 0.661). The deceleration ability of heart rate based on the PRSA algorithm can eliminate non-periodic components, suffer less interference than the traditional HRV index, and have a stronger ability to predict adverse cardiovascular events [12]. After adjusting for confounding factors such as age, sex, ASA physical status, albumin level, BMI, and baseline MAP, DC could still be used as an independent predictor of PIH, suggesting that DC has a broader application space than traditional HRV.

Traditionally, preoperative hypovolemia is considered as the leading cause of hypotension after induction. However, the study by Kahn et al. concluded that fluid optimization could not reduce the incidence of hypotension after induction [8]. In fact, with the development of the concept of enhanced recovery after surgery (ERAS), patients have rarely experienced true preoperative hypovolemia. Moreover, unthinkingly carrying out fluid optimization without paying attention to patients’ preoperative autonomic nerve function cannot influence the cardiac afterload and cannot effectively improve cardiac output and effective circulating blood volume [7]. The dynamic regulation of cardiac output, perfusion pressure, and volume distribution in normal physiology depends on the autonomic nervous control system [11]. In the study of Hanss et al., LF/HF of patients with severe hypotension risk did not improve even with standardized volume loading [28], which emphasizes the importance of autonomic nerve function in perioperative patients. Ferrario et al. had found an interesting phenomenon that indicators of heart rate variability are related to central blood volume [29]. This suggests a noteworthy link between autonomic function and effective blood volume. Effective blood volume was not assessed using ultrasound equipment in this trial, and it is not known whether there is a direct link between DC and blood volume. More studies may be needed to explore this in the future.

In this study, the PIH rate was 45.2%, consistent with the rate observed at our past study (50%) and reported by Wang et al. (48.6%). In this trial, we defined the induction period as 10 min after induction and intubation to avoid large changes in blood pressure caused by short-term stimulation and to obtain continuous and stable blood-pressure data. Bijker et al. summarized 140 definitions of IOH [30], and we chose hypotension to be characterized as either a reduction in MAP exceeding 30% from the initial baseline or a MAP value below 65 mmHg sustained for at least one minute A threshold of 65 mmHg was chosen to ensure patient safety better and reduce hypotension damage to vital organs [31, 32]. At the same time, the 1-min interval avoids large short-term changes in blood pressure due to sensor malfunction or improper operation. The study by Kho et al. pointed out that the time course (e.g., the rate of blood pressure decline during induction) is rarely included as an observational measure in the assessment of hemodynamic instability [33]. In the future, researchers may need to develop a more scientific definition of PIH for specific research populations to explore the association between DC and hypotension after induction.

This study has several limitations. First, our inclusion criteria were broad, and the absolute threshold of MAP was used to define PIH. The results may not be generalizable to specific surgical procedures or particular patient populations. Secondly, although DC is less confounded than traditional HRV indicators, factors such as mental stress and anxiety may still cause inaccurate measurement [12]. More accurate data may be obtained by including indicators such as mental stress and preoperative anxiety of patients in the trial measures. Finally, a more significant decrease in MAP was found in patients in the PIH group in this trial, but it cannot be concluded whether DC is associated with the decrease in MAP. This was due to the exclusion of patients with hypertension, the difference in baseline MAP between the two groups, and the need for immediate rescue measures if a patient had a significant decrease in MAP. These all lead to a limited reduction in MAP, which may be greater in the PIH group, and more studies may be needed to explore the association in the future.

Conclusions

DC measured by ECG 5 min before anesthesia can predict PIH to some extent in patients with ASA status I or II. Compared with RMSSD and LF/HF, DC had greater power to predict PIH as an independent predictor.

Abbreviations

ASA

American Society of Anesthesiologists physical status

DC

Deceleration capacity

ECG

Electrocardiogram

PIH

Post-induction hypotension

IOH

Intraoperative hypotension

RMSSD

Root mean square density

LF/HF

Low frequency/high frequency ratio

MAP

Mean arterial pressure

HRV

Heart rate variability

ROC

Receiver operating characteristic

AUC

Area under the curve

RR Interval

The time interval between two adjacent R waves on ECG

PRSA

Phase-rectified signal averaging

LFnu

Low frequency normalized units

HFnu

High frequency normalized units

FFT

Fast Fourier transform

LF

Low frequency

HF

High frequency

VLF

Very low frequency

HR

Heart rate

BMI

Body mass index

Authors’ contributions

Hongwei Qi, Maofeng Shi, Manlin Duan, Tao Sun, Miaomiao XU and Kangli Hui contributed to the study’s conception and design. Manlin Duan supervised this study. Hongwei Qi, Maofeng Shi and Manlin Duan performed material preparation and data collection. Hongwei Qi and Maofeng Shi performed data analysis. The first draft of the manuscript was written by Hongwei Qi. Tao Sun, Kangli Hui, Teng Wu, Miaomiao XU and Manlin Duan commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

None.

Data availability

The datasets used and/or analyzed during the current study are available from the corre-sponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was performed using the principles of the Declaration of Helsinki. The ethics committee of the Jinling Hospital provided ethical approval for this study (2024DZKY-092–02) on November 25, 2024. This study was officially registered in the Chinese Clinical Trial Registry with registration number ChiCTR2400094595 (registration date: 25/12/2024). Written informed consent was obtained from all eligible patients.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.