Carotid artery velocity-time integral variation combined with passive leg raising test to predict hypotension after induction in older patients under general anesthesia: a prospective study

Abstract

Background

Older patients are at an increased risk of developing hypotension following the induction of general anesthesia, which is linked to a higher incidence of postoperative complications, mortality, and morbidity. This study aimed to investigate the effectiveness of carotid velocity time integral variation (ΔcVTI) combined with the passive leg raising test (PLR) in predicting hypotension after anesthesia induction in elderly patients.

Methods

This prospective observational study enrolled 75 older patients (65–75 years, ASA II–III) undergoing elective surgery under general anesthesia. Carotid blood flow was continuously monitored using a wearable Doppler ultrasound patch, and ΔcVTI (%) was calculated during passive leg raising. Anesthesia was induced with etomidate, alfentanil, and rocuronium following a standardized protocol. Post-induction hypotension was defined as mean arterial pressure < 65 mmHg, a > 20% reduction from baseline, or systolic pressure < 90 mmHg within 3 min after induction.

Results

The incidence of hypotension observed in the study was 29 cases (39%). The area under the ROC curve for ΔcVTI in predicting hypotension after anesthesia induction was 0.84 (95% CI, 0.74 to 0.94; P < 0.001), with an optimal cutoff value of 13.6%, a sensitivity of 72.4% (95% CI, 54.3–85.3%), and a specificity of 97.8% (95% CI, 88.7–99.6%). Logistic regression analysis identified ΔcVTI as the sole independent risk factor for hypotension following anesthesia induction.

Conclusions

ΔcVTI combined with the preoperative passive leg raising test may serve as a simple, non-invasive, and reliable method for predicting anesthetic hypotension in older patients.

Trial registration

Clinical Trial Registry on January 8, 2025. (www.chictr.org.cn; ChiCTR2500095534).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12877-026-07267-3.

Introduction

Post-induction hypotension (PIH) is a common clinical phenomenon of hemodynamic instability after general anesthesia, particularly in older patients with reduced physiological reserves and multiple comorbidities [1, 2]. PIH has been independently associated with increased risks of acute kidney injury, myocardial injury, stroke, and postoperative delirium after non-cardiac surgery [3–5]. Emerging evidence suggests that even brief episodes of hypotension can result in adverse outcomes, underscoring the importance of early identification and prevention [6, 7]. In contrast to intraoperative hypotension, which can arise from multiple factors, the primary causes of post-induction hypotension are the preoperative status of the patient and the anesthetic medications administered. Consequently, this condition is both predictable and preventable.

One key modifiable contributor to PIH is preoperative intravascular volume status. Relative hypovolemia — due to fasting, bowel preparation, or pre-existing disease — can increase susceptibility to hypotension during anesthetic induction [8, 9]. Therefore, assessing fluid responsiveness prior to induction may aid in early risk stratification and optimization.

Recently, Doppler ultrasound of the common carotid artery has gained widespread use for assessing volume status and predicting fluid responsiveness in various clinical scenarios [10–12]. It advantages include non-invasiveness, a larger carotid artery diameter, a superficial location, and good reproducibility [13]. Notably, there is evidence that the carotid artery velocity time integral can serve as a reliable index for predicting fluid responsiveness [11, 14]. However, no studies have specifically evaluated its predictive value for hypotension following the induction of anesthesia in older patients. Therefore, this study aimed to investigate the predictive value of the carotid artery velocity time-integral variation for the occurrence of hypotension after induction in older patients undergoing general anesthesia.

Methods

study population

This study received approval from the Ethics Committee of the First Affiliated Hospital of Zhengzhou University, Henan, China(2024-KY-1217-002) and was registered with the Chinese Clinical Trial Registry (www.chictr.org.cn; ChiCTR2500095534). Conducted between January and March 2025, the study involved elderly patients scheduled for elective surgery under general anesthesia at our hospital. Written informed consent was obtained from all participants at least 24 h prior to surgery.

We recruited older patients aged 65 to 75 years with American Society of Anesthesiologists (ASA) physical status II-III who were scheduled for elective surgery under general anesthesia. The exclusion criteria included: a body mass index (BMI) lower than 18 kg/m² or higher than 30 kg/m²; diabetes; coronary artery disease; heart conditions (which encompass cardiomyopathy and moderate to severe valvular disorders); pulmonary hypertension; preoperative hypertension (systolic blood pressure ≥ 180 mmHg or diastolic blood pressure ≥ 110 mmHg ); atherosclerosis of the aorta and peripheral arteries; preoperative cervical vascular ultrasound irregularities (covering all levels of stenosis and anatomical variations); as well as individuals with a previous history of neck surgery or trauma.

Carotid doppler monitoring

All patients fasted after midnight (NPO) without bowel preparation. Upon entering the anesthesia preparation room, standard monitoring was initiated, including three-lead electrocardiography (ECG), non-invasive blood pressure, and pulse oximetry. After resting quietly for at least 5 min, patients were positioned in a semi-recumbent posture with the trunk elevated 45° (lower limbs horizontal) as the initial position.

A wearable ultrasound patch was employed to measure carotid velocity-time integral (cVTI). The ultrasound patch was placed on the left side of the patient’s neck at the plane of the inferior border of the thyroid cartilage, perpendicular to the trachea (Fig. 1a). The probe was moved laterally until the strongest carotid Doppler audiovisual signal was recorded on the user interface, then properly secured and connected to the CADFlow™ series UCM Doppler ultrasound system (Sensus Medical, Suzhou, China) (Fig. 1b). cVTI represents the area under the carotid blood flow velocity-time curve (unit: cm), with the device outputting cVTI values for the current cardiac cycle every second.

Fig. 1.

The wearable Doppler ultrasound device. a The apparatus positioned around the neck of a participant. b The visual interface of the ultrasound system shown on the device

Patients were instructed to remain motionless in the initial position for at least three minutes. Baseline cVTI was calculated as the average of all valid cVTI values during the final 10 s before PLR (validity criteria: complete Doppler spectrum). Subsequently, the torso was synchronously lowered to the horizontal position while elevating both legs to 45°, with this maneuver completed within ≤ 5 s. cVTI60-90s was calculated by taking the mean of all cVTI values during the t = 60–90  second period after defining leg elevation initiation as t = 0.

After testing completion, patients were returned to the initial position while continuing observation to verify carotid velocity-time integral recovery to baseline. To avoid sympathetic stimulation effects, we required that the average heart rate change between the baseline period (final 10 s before PLR) and PLR period (60–90 s) not exceed 5% [15]. When exceeding this threshold, passive leg raise testing was repeated after at least a 5-minute “washout period”. Following completion of Doppler measurements, patients were transferred to the operating room to begin anesthetic induction.

Anesthesia induction

Upon entering the operating room, patients received oxygen via face mask at a flow rate of 2 L/min. Standard monitoring was applied, including systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate (HR), oxygen saturation (SpO₂), and electrocardiography (ECG) (Mindray Biomedical Electronics Co., Shenzhen, China). Intravenous access was established, and a routine infusion of compound sodium chloride was initiated at a rate of 10 mL/kg/h.

General anesthesia was induced with intravenous etomidate (0.2 mg/kg), alfentanil (50 µg/kg), and rocuronium (0.6 mg/kg). Tracheal intubation was performed 3 min after the administration of these agents. Anesthetic management was conducted under the supervision of the attending anesthesiologist.

Data collection

Baseline systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), and heart rate (HR) were obtained by averaging triplicate readings taken at 3-minute intervals while the patient remained calm prior to induction. SBP, DBP, MAP, and HR were subsequently recorded at 1, 2, and 3 min post-induction, with the lowest values being noted. PIH was defined as either an absolute MAP below 65 mmHg, a drop in MAP exceeding 20% from baseline, or an absolute systolic blood pressure below 90 mmHg within three minutes of anesthesia induction. In cases of hypotension, norepinephrine was administered intravenously at a dose of 4 µg, with the treatment repeated if necessary. Significant bradycardia, defined as a heart rate lower than 40 beats per minute, was treated with an intravenous bolus of atropine (0.3 mg), and the nadir value was recorded.

Sample size calculation and statistical analysis

The required sample size was calculated using G*Power software (version 3.1.9.2). A pilot study involving 30 subjects indicated a 40% incidence of post-induction hypotension in older patients. We proposed that the predictive validity of the ΔcVTI regarding post-induction hypotension in older patients might be lower, approximated at 0.70. Consequently, calculations for sample size indicate a necessity for at least 73 individuals to identify a 0.20 difference in the ROC curve for the ΔcVTI (0.70) versus the null hypothesis (0.50), maintaining a power of 0.85 and a two-tailed type I error rate of 0.05. To account for a possible dropout rate of 10%, our recruitment target was set at 82 patients.

Statistical analyses were performed using SPSS (version 26.0) and GraphPad Prism (version 8.0) software. Continuous variables were tested for normality (Shapiro–Wilk test) and expressed as mean ± SD or median (IQR), while categorical variables were summarized as counts and percentages. Between-group comparisons were made using the t-test or Mann–Whitney U test for continuous variables and the Chi-square or Fisher’s exact test for categorical variables. Within-group changes in hemodynamics were analyzed with paired tests. The predictive value of ΔcVTI was assessed using ROC curve analysis, with AUC, 95% CI, optimal cutoff (Youden index), and gray-zone analysis reported. Independent predictors of hypotension were identified by multivariable logistic regression, with results expressed as odds ratios (OR) and 95% CI. A two-tailed P < 0.05 was considered statistically significant.

Results

A total of 82 patients were evaluated; however, 1 patient was excluded due to an entry systolic blood pressure exceeding 180 mmHg, 3 patients had invalid ultrasound measurements, and 3 patients exhibited heart rate changes greater than 5% following PLR. Consequently, 75 patients completed the study, and no data were lost. (Fig. 2). According to Table 1, of the 75 patients, 29 developed hypotension after anesthesia induction and 46 did not. Baseline characteristics, including age, gender, BMI, hypertension history, ASA status, baseline HR, SBP, DBP, MAP, and cVTI, as well as induction doses of etomidate, alfentanil, and rocuronium, showed no significant differences between groups ( P > 0.05). The distribution of surgical types also showed no significant differences. The only significant difference was in ΔcVTI, which was markedly higher in the hypotension group (17.50% [8.8–26.5%]) than in the non-hypotension group (1.94% [− 7.3–9.3%], P < 0.001).

Fig. 2.

Flow chart of the study

Table 1.

Subject demographics, clinical baseline values and carotid ultrasound parameters

Parameter	Hypotension (n = 29)	Non-hypotension (n = 46)	P -Value
Age (years)	67.79 ± 3.19	67.43 ± 2.52	0.590
Gender (Male/Female)	11/18	22/24	0.401
BMI (kg/m2)	23.81 ± 2.51	24.24 ± 3.03	0.527
Baseline cVTI	22.81 ± 6.10	25.30 ± 4.66	0.067
ΔcVTI (%)	17.50[8.8–26.5]	1.94[-7.3-9.3]	< 0.001*
Hypertension (with/without)	6/29	14/46	0.353
ASA(II/III), n(%)
II	25(86.21)	40(86.96)	0.926
III	4(13.79)	6(13.04)	0.926
Baseline HR(beats/min)	72.52 ± 9.04	69.59 ± 8.50	0.094
Baseline Systolic BP (mmHg)	144.14 ± 12.44	141.83 ± 12.57	0.439
Baseline Diastolic BP (mmHg)	81.52 ± 8.72	79.13 ± 7.76	0.234
Baseline MAP (mmHg)	102.39 ± 8.10	100.03 ± 6.47	0.167
Etomidate (mg)	15.82 ± 2.33	16.05 ± 2.70	0.690
alfentanil (mg)	3.2 ± 1.6	3.5 ± 1.8	0.098
rocuronium (mg)	57.6 ± 10.3	63.0 ± 11.2	0.099
Gynecological Surgery (n, %)	2 (6.9%)	2 (4.3%)	0.638
Hepatobiliary Surgery (n, %)	4 (13.8%)	5 (10.9%)	0.727
Orthopedic Surgery (n, %)	6(20.7%)	6 (13.0%)	0.519
Thyroid Surgery (n, %)	1 (3.4%)	0 (0%)	0.387
Urologic Surgery (n, %)	1 (3.4%)	3 (6.5%)	1.000
Breast Surgery (n, %)	2 (6.9%)	0 (0.0%)	0.146
Neurosurgery (n, %)	2 (6.9%)	4 (8.7%)	1.000
Thoracic Surgery (n, %)	11 (37.9%)	26 (56.5%)	0.156

Data are expressed as number of patients (n) and percentage (%) for categorical variables; Other variables are reported as mean ± standard deviation or median (Q1-Q3) depending on the distribution type

ASA American Society of Anesthesiology, cVTI carotid velocity time integral, MAP Mean arterial pressure

*Statistically significant p<0.05

Figure 3 displays the distribution of ΔcVTI values using boxplots with jittered individual data points, stratified by hypotension status. The median ΔcVTI was 17.50% (interquartile range [IQR] 8.8–26.5%) in the hypotension group (n = 29) and 1.94% (interquartile range [IQR] -7.3-9.3%) in the non-hypotension group (n = 46). A Mann-Whitney U test demonstrated a statistically significant difference between groups (U = 213.00, Z=-4.765, P < 0.001), indicating that the hypotension group had significantly higher ΔcVTI values compared to the non-hypotension group.

Fig. 3.

Distribution of ΔcVTI in hypotension and non-hypotension groups. *** p<0.001

Baseline MAP was similar between groups (non-hypotension: 100.03 ± 6.47, n = 46; hypotension: 102.39 ± 8.10, n = 29). Post-induction MAP was lower in the hypotension group than in the non-hypotension group (68.71 ± 7.80 vs. 85.79 ± 8.36; P < 0.001). The decrease from baseline was greater in the hypotension group (ΔMAP = 33.68 ± 9.25) than in the non-hypotension group (ΔMAP = 14.24 ± 7.47; P < 0.001) (Fig. 4a).

Fig. 4.

MAP and HR before and after induction in two groups. MAP: mean arterial pressure; HR: heart rate

Baseline HR did not differ between groups (69.59 ± 8.50 vs. 72.52 ± 9.04; P = 0.094). Post-induction HR was lower in the hypotension group than in the non-hypotension group (60.59 ± 10.61 vs. 65.33 ± 10.63; P < 0.05). The decrease from baseline was greater in the hypotension group (ΔHR = 11.93 ± 10.82) than in the non-hypotension group (ΔHR = 4.26 ± 10.22; P = 0.003) (Fig.4b).

The ability of ΔcVTI to predict post-induction hypotension is illustrated in the accompanying Fig. 5; Table 2. The area under the ROC curve (AUC) for ΔcVTI was 0.84 (95% CI, 0.74–0.94; P < 0.001). The optimal cutoff value was 13.6%, which yielded a sensitivity of 72.4% (95% CI, 54.3–85.3%) and a specificity of 97.8% (95% CI, 88.7–99.6%), with a maximum Youden index of 0.702. The grey zone was identified between − 3.7% and 12.4% (20.3% of patients).

Fig. 5.

ROC curve of ΔcVTI predicting the occurrence of post-induction hypotension in older patients with general anesthesia ROC: Receiver Operating Characteristic; cVTI: carotid velocity time integral

Table 2.

Diagnostic performance of ΔcVTI to predict postinduction hypotension

Parameter	AUROC curve(95%CI)	P- Value	Optimal cutoff value	Gray zone	Patients in gray zone(%)	Sensitivity(%) (95%CI)	Specificity(%) (95%CI)	Youden Index J
ΔcVTI	0.84 (0.74 to 0.94)	< 0.001*	13.6	-3.7 to 12.4	20.3	72.4 (54.3to85.3)	97.8 (88.7to99.6)	0.702

AUROC The area under the Receiver Operating Characteristic curve, cVTI carotid velocity time integral

*statistically significant p<0.05

This binary logistic regression model was designed to identify the risk factors influencing the occurrence of hypotension following anesthesia induction in older patients (Table 3). The model included age, gender, BMI, ASA score, baseline MAP, baseline SBP and ΔcVTI. The analysis revealed that the only independent determining factor was ΔcVTI (odds ratio [OR] = 1.149; 95% confidence interval [CI]: 1.071–1.233; P < 0.001). In older patients, the risk of developing post-induction hypotension increased by 14.9% for each unit increase in ΔcVTI.

Table 3.

Binary Logistic Regression Model for Risk Factors of Hypotension After General Anesthesia Induction

Predictors	Reggression Coefficient	Odds Ratio (95% CI)	P-value
Age	-0.045	0.956 (0.765-1.194)	0.691
Gender	-0.373	0.689 (0.178-2.670)	0.590
BMI	-0.137	0.872 (0.689-1.104)	0.256
ASA	-0.554	1.741 (0.217-13.955)	0.602
Hypertension	-1.466	0.231 (0.046-1.157)	0.075
Baseline SBP	0.039	1.040 (0.962-1.124)	0.328
Baseline MAP	0.009	1.009 (0.891-1.142)	0.887
ΔcVTI	0.139	1.149 (1.071-1.233)	0.000*

ASA American Society of Anesthesiologists, cVTI carotid velocity time integral, MAP Mean arterial pressure, SBP systolic blood pressure

*statistically significant p<0.05 

Discussion

In this study, we found that the variation of the common carotid artery velocity-time integral measured during the passive leg-raising test prior to the induction of anesthesia serves as a reliable predictor for the occurrence of hypotension following the induction of general anesthesia in older patients, demonstrating an area under the receiver operating characteristic (AUROC) value of 0.84 and an optimal cutoff value of 13.6%. Specifically, a ΔcVTI exceeding 13.6% prior to anesthesia induction indicates a significantly heightened risk of post-induction hypotension. Consequently, we propose that the combination of ΔcVTI and PLR represents an effective approach for predicting post-induction hypotension in older patients undergoing general anesthesia.

In our cohort, 39% of patients developed hypotension following anesthetic induction, which is lower than the incidence reported in previous studies [16]. Earlier investigations have established advanced age as an independent risk factor for PIH [17, 18]. The lower mean age of our study population compared with prior studies may partly explain this discrepancy. In addition, variations in PIH incidence could be attributed to the absence of a standardized definition, as different criteria may yield divergent rates [17]. The definition of the post-induction period may also influence the reported incidence; in our study, measurements were terminated before tracheal intubation, given that direct laryngoscopy and intubation can induce sympathetic stimulation and hemodynamic fluctuations [19].

Anesthetics such as etomidate [20] and alfentanil [21] exert cardiovascular depressant effects, leading to reductions in mean arterial pressure and heart rate. Post-induction hypotension may therefore result from preexisting hypovolemia compounded by the synergistic hemodynamic effects of these agents. In our study, patients who developed hypotension after anesthetic induction exhibited significantly higher ΔcVTI during the passive leg-raising test compared with those who did not. This finding suggests that the extent of heart rate and blood pressure reduction may correspond to the intravascular volume status prior to induction [22]. Older patients are particularly susceptible to hemodynamic instability and hypotension due to a higher prevalence of left ventricular diastolic dysfunction, reduced vascular compliance and reactivity, increased sensitivity to anesthetic agents, and prolonged preoperative fasting and preparation [23]. These factors collectively contribute to preoperative intravascular volume depletion and may further exacerbate the risk of hypotension following anesthetic induction. Therefore, careful assessment of preoperative volume status is essential, particularly when other contributing factors cannot be modified. The results of our logistic regression analysis revealed that among variables such as gender, BMI, ASA score, history of hypertension, ΔcVTI, baseline SBP and MAP, only a high ΔcVTI emerged as an independent risk factor for post-anesthesia hypotension, underscoring the predictive value of ΔcVTI in the context of general anesthesia.

Cardiac ultrasound combined with PLR has been used to assess fluid responsiveness [24, 25]. PLR is a simple and noninvasive method that transiently increases venous return by shifting blood from the lower extremities to the intrathoracic compartment [25]. In recent years, carotid artery ultrasound has also been increasingly utilized to evaluate fluid responsiveness. Marik experimentally demonstrated that carotid artery velocity-time integral serves as a promising surrogate parameter for cardiac output, particularly when employing the leg-raising test to predict fluid responsiveness [24]. A prospective observational study also demonstrated good agreement between carotid and aortic velocity–time integral, with a Cohen’s kappa coefficient of 0.84 (95% CI, 0.68–0.99) [11]. The superficial location of the carotid artery facilitates easy acquisition of ultrasound measurements by anesthetists.

A recent study by Cheong et al. demonstrated that carotid ΔVTI in combination with PLR provides an effective non-invasive method for predicting fluid responsiveness in critically ill patients, with an AUC of 0.869 (95%CI 0.743–0.947) for identifying fluid responders [14]. Notably, Dai conducted a fluid replenishment experiment and found that the area under the receiver operating characteristic curve for carotid artery ΔVTI in predicting fluid responsiveness among spontaneously breathing parturients was 0.821 (95% CI, 0.720–0.922) [12]. This underscores the relevance of using cVTI to assess volume responsiveness.

There is limited literature on the application of ΔcVTI combined with PLR in older patients. Zieleskiewicz reported an area under the ROC curve (95% CI) of 0.8 (0.6–0.9; p < 0.001) using aortic VTI combined with PLR to predict hypotension in spontaneously breathing mothers after spinal anesthesia, with an optimal cutoff value of 8% [26]. In comparison, our study in an elderly population under general anesthesia found a higher predictive value for ΔcVTI, with an AUC of 0.84 (95% CI, 0.74–0.94; p < 0.001) and an optimal threshold of 13.6%.

Several protocol differences exist between our study and Zieleskiewicz’s. First, we conducted carotid artery ultrasound measurements specifically in an elderly population undergoing general anesthesia. Secondly, we ensured that the patient’s heart rate change remained less than 5% before and after the leg elevation, and we monitored whether the patient’s cVTI returned to baseline levels following PLR.

As demonstrated in earlier studies, accurate VTI assessment requires averaging over ≥ 2–3 full respiratory and multiple cardiac cycles to minimize physiologic variability [27, 28]. At typical respiratory and heart rates, a 10-second Doppler acquisition window is generally sufficient [29]. Given that PLR induces a brief and reversible preload challenge, with peak effects within 30–90 s, we chose a 30-second window to capture the response while minimizing noise from postural or autonomic changes [30, 31]. To minimize autonomic confounding, individuals with marked heart rate variability during PLR were excluded. Reliable interpretation of dynamic indices such as PPV, SVV, and PLR requires regular rhythm, stable ventilation, and relatively steady heart rate [32–34]. Excessive heart rate fluctuation may independently affect stroke volume and pressure variability, thereby distorting the ΔVTI response [35].

The strengths of our study lie in the fact that it is the first to evaluate the predictive value of ΔcVTI combined with PLR for post-induction hypotension. Additionally, our research utilized a hands-free carotid Doppler patch, which facilitates continuous quantitative monitoring of carotid blood flow while minimizing the effects of motion. However, this study acknowledges several limitations. First, the association between carotid and left ventricular outflow tract VTI during PLR may be attenuated by mechanisms such as vasodilation, blood redistribution, and alterations in flow velocity profiles [36, 37]. Second, the induction regimen utilized in our study involved etomidate, which may limit the applicability of our findings to older patients undergoing propofol induction. Finally, this study did not focus on specific types of surgeries, and the varying disease states of patients undergoing different surgical procedures may introduce bias into the results. Future work should further validate our findings by enrolling larger and more homogeneous patient cohorts, exploring different induction regimens, and assessing whether pre-induction intravenous fluid administration can reduce ΔcVTI and thereby decrease the incidence of post-induction hypotension.

Conclusions

In summary, ΔcVTI measured combined with the passive leg-raising test has the potential to predict hypotension after induction in older patients undergoing general anesthesia, thereby assisting in guiding individualized treatment. Nonetheless, our findings require further verification in subsequent studies.

Abbreviations

ASA

American Society of Anesthesiologist

BMI

Body Mass Index

CI

Confidence interval

SBP

Systolic blood pressure

DBP

Diastolic blood pressure

MAP

Mean arterial pressure

AUROC

Area under the receiver operating characteristic

ΔcVTI

Variation of the common carotid artery velocity-time integral

cVTI

Carotid velocity time integral

PLR

Passive leg raising test

IVCCI

Inferior vena cava collapse index

PPV

Pulse pressure variation

SVV

Stroke volume variation

Authors’ contributions

Conception and design: Zhiwei Ge, Feifei Qin, Zhongming Lian, Yaru Huang, Jianjun Yang, Dandan Tian. Data collection: Zhiwei Ge, Feifei Qin, Boyan Zhang. Data analysis: Zhongming Lian, Yaru Huang. Drafting the manuscript: Zhiwei Ge. The final manuscript was reviewed and approved by all authors.

Funding

This work was supported by China Primary Health Care Foundation (YLGX-WS-2020005), Beijing Medical Award Foundation (YXJL-20210307-0677), and Henan Provincial Science and Technology Research Project (242102311101).

Data availability

The data sets utilized and/or examined in this study can be obtained from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

For the recruited participants, informed consent forms in writing duly were collected before their examination. This research was carried out in alignment with the principles outlined in the Declaration of Helsinki and received approval from the Ethics Committee at the First Affiliated Hospital of Zhengzhou University in China.

Competing interests

The authors declare no competing interests.