External validation of CLOTS-AF score in patients with atrial fibrillation undergoing transesophageal echocardiography

Abstract

Background

Predicting thromboembolic events caused by left atrial appendage thrombosis (LAAT) is crucial since it is the most concerning complication of nonvalvular atrial fibrillation (NVAF).

Objective

The objective of this study was to establish the predictive effectiveness of a novel scoring system known as the CLOTS-AF score in identifying the existence of thrombus using transesophageal echocardiography (TEE) in patients with NVAF and to compare it with the CHA₂DS₂-VASc score.

Methods

This study was conducted retrospectively with 668 NVAF patients who underwent TEE in a single centre. The patients were categorized into two categories based on the presence or absence of LAAT on TEE. The CLOTS-AF score was computed for all patients, which was then compared to the CHA₂DS₂-VASc score.

Results

Thrombus was detected in the left atrial appendage (LAA) of 114 (17%) individuals. The study revealed that the thrombus (+) group had substantially higher CHA₂DS₂-VASc scores and CLOTS-AF scores (3 (2–4); 4 (3–6), p < .001), respectively. Furthermore, the CLOTS-AF score was shown to be a better predictor of LAAT than the CHA₂DS₂-VASc score (AUC: 0.70 vs. AUC: 0.61, p < .001).

Conclusions

This study could validate the CLOTS-AF score in an external cohort by demonstrating its significant predictive power and underscoring its vital role in clinical practice.

1. Introduction

Atrial fibrillation (AF), the most prevalent supraventricular arrhythmia, is characterized by erratic electrical activity in addition to ineffective contractions of the atria. AF is present in 2% of the general population, although its occurrence becomes more common as people get older, with a prevalence of 10–12% in individuals over 80 years old [1]. The most feared complication of non-valvular AF (NVAF) is the occurrence of thromboembolic (TE) events caused by thrombosis in the left atrial appendage (LAA) [2].

Ischemic strokes account for approximately 85% of all strokes, with significant causes including 25% resulting from cerebral small vessel arteriolosclerosis, another 25% stemming from cardioembolism, and 20% arising from large artery atherothromboembolism. It is essential to recognize that cardioembolic strokes are primarily triggered by AF, but they can also occur due to infective endocarditis, paradoxical embolism linked to a patent foramen ovale, or even an atrial tumour [3]. Notably, in cases of NVAF, an astonishing 90% of embolic events are attributed to left atrial appendage thrombus (LAAT) [4]. Understanding these causes can significantly impact prevention and treatment strategies.

The prevention of strokes is a significant focus in the treatment of AF [5]. Anticoagulation therapy with vitamin K antagonists (VKAs) or direct oral anticoagulants (DOACs) effectively decreases the incidence of stroke and mortality [6]. Patients who require restoration of sinus rhythm and are without sufficient anticoagulant medication should have transesophageal echocardiography (TEE) and other exams to exclude LAAT [7].

The current guidelines recommend using the CHA₂DS₂-VASc scoring system to assess the risk of TE in patients with NVAF. In this scoring system, patients with a score of ≥2 points in males or ≥3 points in females are considered to have a high risk of TE and are likely to see significant benefits from anticoagulation therapy [8]. The CHA₂DS₂-VASc score is intended to forecast the risk of TE and stroke, rather than evaluate the risk of LAAT, and demonstrated inadequate predictive capability in studies focused on LAAT risk assessment [9,10]. The CLOTS-AF score is a novel scoring system designed to evaluate the risk of LAAT rather than TE and stroke, in contrast to the CHA₂DS₂-VASc score. This comprehensive tool assesses various significant factors, including creatinine levels >1.5 mg/dL (2 points), a left ventricular ejection fraction (LVEF) <50% (2 points), left atrial volume index (LAVI) >34 mL/m² (indicating overload, 1 point), tricuspid annular plane systolic excursion (TAPSE) <17 mm (2 points), history of prior stroke (3 points) and presence of AF rhythm (2 points). The score can range from 0 to 12 points, providing a nuanced understanding of a patient’s risk [11]. However, it is important to note that a definitive cut-off score for identifying LAAT in different populations risk has yet to be established, highlighting the need for further studies and application in clinical practice.

The incidence of LAAT in patients anticoagulated per current recommendations exhibits considerable diversity across various research, prompting essential inquiries about the dependability of the CHA₂DS₂-VASc grading method [12–16]. This study can serve as an external validation of the CLOTS-AF score by evaluating its capacity to predict LAAT in a novel cohort, thereby expanding upon the findings of Segan et al. [11]. Consequently, this study aims to demonstrate the predictive efficacy of a novel scoring system, the CLOTS-AF score, in detecting the presence of thrombus with TEE in patients with NVAF who are scheduled for cardioversion or ablation procedures to restore sinus rhythm.

2. Materials and methods

2.1. Study design and participants

This present study was a retrospective analysis conducted at a single centre. The review encompassed TEE data conducted between January 2018 and January 2021. Only TEE examinations performed within two days leading up to the operation date for patients slated for catheter ablation were considered for inclusion. Additionally, TEE was performed in patients undergoing cardioversion when AF had lasted more than 48 h or when patients were not using DOACs at appropriate dosages, or when their international normalized ratio (INR) fell outside the recommended range due to VKA administration.

The exclusion criteria were adults below the age of 18, those with more than moderate mitral stenosis, those who had post-mitral valve repair and post-mechanical or biological valve replacement, those with congenital heart disease, those who had heart transplantation or left ventricular assist device (LVAD), and patients suffering from acute infections and autoimmune diseases. After implementing these specific criteria, a total of 668 cases out of the initial 873 were selected for the final study. A comprehensive analysis of the electronic medical database was performed to gather information on demographic, clinical and echocardiographic factors.

The study was conducted according to the principles of the Declaration of Helsinki and its subsequent revisions. This study was approved by the local ethics committee.

2.2. Echocardiography

Transthoracic echocardiography (TTE) and TEE procedures were performed with the EPIQ CVx cardiovascular ultrasound system and X5-1 and X8-2t transducers (PHILIPS, Cambridge, MA). All measurements complied with the norms established by the American Society of Echocardiography guidelines [17,18]. Both TEE and TTE were performed and analysed by experienced cardiologists specialized in echocardiography and offline images of TEE exams to ascertain the presence or absence of thrombus. Warfarin was not discontinued on the day of the procedure. Patients prescribed with DOACs were instructed to abstain from taking their morning dosage on the day of the medical procedure. If there were no instances of bleeding, it was advised to recommence taking the medication in the evening of the same day. The same anticoagulation directives were implemented during the ablation operation.

The existence of thrombus in the LAA was evaluated using several perspectives, such as three-dimensional datasets and the multiplanar reconstruction of these datasets. A thrombus was described as a clearly defined, round or irregular, uniform-density echogenic image in the left atrium (LA) that was different from the density of the surrounding myocardial tissue.

The LAA sludge was identified as a formless, thick and constantly changing echogenic density that did not allow flow to penetrate into the mass during Doppler investigation. Dense spontaneous echo contrast (SEC) is characterized by a swirling echogenic density that is present throughout the cardiac cycle. It may or may not have a viscous look, known as sludge. This condition occurs in the absence of a distinct mass and can be identified by utilizing suitable gain settings to avoid artefacts. The Fatkin classification categorizes. Dense SEC is graded 3+ and 4+ based on severity levels from 0 to 4 + [19]. The velocity of LAA emptying was measured by positioning the pulse-wave Doppler gate within 1 cm of the LAA opening. A reduced emptying velocity was defined as a value below 0.4 m/s.

Baseline TTE measurements were considered valid if they were conducted within 12 months prior to the TEE date. LVEF was determined using the modified Simpson method for TTE measures. Additionally, LA was assessed from the parasternal long-axis view. TAPSE was assessed using the apical four-chamber view. Mitral regurgitation was evaluated with semiquantitative and quantitative techniques as outlined in the applicable standards.

2.3. Definitions

AF is described by the absence of P waves and irregular RR intervals on a typical 12-lead ECG or 24-hour rhythm monitoring for more than 30 s. Paroxysmal AF is recognized as an arrhythmia that ends on its own or with medical intervention within a period of seven days after its inception. However, persistent AF is defined as an arrhythmia that lasts for more than seven days, including episodes that are ended by cardioversion (either through medications or electrical cardioversion) after more than seven days [8].

The CHA₂DS₂-VASc score was assessed for CHF (1 point), HT (1 point), age (2 points for ages ≥75 years), DM (1 point), previous stroke/TIA (2 points), vascular disease (including peripheral vascular disease and myocardial infarction (MI)) (1 point), age (1 point for ages 65–74 years) and sex category (1 point for women). A score of 0 for men or 1 for women was classified as low risk, a score of 1 for men or 2 for women was classified as moderate risk, and ≥2 for men or ≥3 for women was considered high risk [20].

The CLOTS-AF risk score comprises several factors: creatinine levels >1.5 mg/dL (2 points), LVEF <50% (2 points), LAVI >34 mL/m² (indicating overload, 1 point), TAPSE <17 mm (2 points), history of prior stroke (3 points) and presence of AF rhythm (2 points). The total score can range from 0 to 12 points [11].

2.4. Statistical analysis

The analyses were carried out using the Statistical Package for Social Sciences, version 20.0 (SPSS; IBM, Armonk, NY). To determine normality, the Kolmogorov–Smirnov test was used. The study group was divided into two groups according to the presence of LAAT. The baseline features, laboratory results, echocardiographic data and anticoagulation therapy of these groups were compared. The median and interquartile range of the allocated individuals’ continuous baseline characteristics were reported, while categorical data were expressed as numbers and percentages. If continuous variables were regularly distributed, they were examined using an independent t-test. The Mann–Whitney U-test was used to examine the parameters that did not have a normal distribution. Categorical data were reported in the form of n (%) and analysed using Chi-square or Fisher’s exact tests. The independent predictors for LAAT were determined using univariable and multivariable Cox regression analysis using enter analysis. Parameters having a p value of <.05 in univariable analysis were included in the multivariate Cox regression analysis. Cut-off values of CLOTS-AF and CHA₂DS₂-VASc scores with the highest sensitivity and specificity were calculated by nonparametric receiver operating characteristics (ROC) curve analysis. The DeLong test was used to compare ROC curves. All data are presented as hazard ratios (HRs) and 95% confidence intervals (CIs). The level of statistical significance was established at p value ≤.05.

3. Results

This study comprised a total of 668 patients. The patients were categorized into two groups based on the presence or absence of a thrombus in the LAA. While no thrombus was observed in the LAA during TEE in 554 (83%) patients, thrombus was observed in the LAA in 114 (17%) patients. CHA₂DS₂-VASc and CLOTS-AF scores were assessed for all patients (Figure 1). The type of AF was mostly paroxysmal (50.9%) in the group without thrombus, while it was long persistent (43.0%) AF in the group with thrombus. There were no notable differences between the two groups in relation to gender and comorbidities, except for smoking, chronic renal failure (CHR) and prior MI. The demographic, clinical features and echocardiographic parameters of the patients were compared based on the presence of thrombus in TEE. The results of this comparison are presented in Table 1. There were no differences between groups in terms of anticoagulation medications that commenced in the peri-procedural period.

Figure 1.

These figures indicate the presence or absence of thrombus in transesophageal echocardiography based on CHA₂DS₂-VASc and CLOTS-AF scores.

Table 1.

Comparison of demographic, clinical characteristics and echocardiographic parameters of the patients according to thrombus presence in transesophageal echocardiography.

	Left atrial appendage thrombus (−), n = 554	Left atrial appendage thrombus (+), n = 114	p Value
Age, years	60 (52–69)	62 (55–69)	.357
Male gender	352 (63.5%)	66 (57.9%)	.257
Hypertension	285 (51.4%)	77 (67.5%)	.002
Diabetes mellitus	132 (23.8%)	41 (36.0%)	.007
Hyperlipidemia	295 (53.2%)	67 (58.8%)	.281
Smoking	114 (20.7%)	42 (37.2%)	<.001
Chronic renal failure	51 (9.2%)	27 (23.7%)	<.001
Chronic obstructive pulmonary disease	74 (14.1%)	10 (8.8%)	.108
Cerebrovascular accident	13 (2.4%)	11 (9.6%)	<.001
Coronary artery disease	144 (26.0%)	40 (35.1%)	.048
Congenital heart disease	26 (4.7%)	5 (4.4%)	.886
Prior myocardial infarction	73 (13.2%)	31 (27.2%)	<.001
Peripheral artery disease	49 (8.8%)	8 (7.0%)	.515
Cardiac device	21 (3.8%)	5 (4.4%)	.790
Atrial fibrillation types
Paroxysmal	282 (50.9%)	22 (19.3%)	<.001
Persistent	162 (29.2%)	43 (37.7%)	.074
Long persistent	108 (19.5%)	49 (43.0%)	<.001
Atrial fibrillation duration, months	12 (2–36)	12 (6–36)
Anticoagulation types
Warfarin	179 (32.3%)	49 (43.0%)	.029
Apixaban	85 (15.3%)	16 (14.0%)	.720
Rivaroxaban	124 (22.4%)	18 (15.8%)	.107
Dabigatran	44 (7.9%)	12 (10.5%)	.378
Edoxaban	45 (8.1%)	7 (6.1%)	.459
None	77 (13.9%)	12 (10.5%)	.322
Echocardiographic parameters
Ejection fraction, %	55 (50–60)	45 (30–55)	<.001
Ejection fraction <50%	136 (24.5%)	58 (50.9%)	<.001
Left atrium anteroposterior, mm	41 (37–46)	44 (40–49)	<.001
Left atrial volume index, mL/m2	41 (37–45)	50.0 (45.0–54.0)	<.001
Left ventricle end-diastolic diameter, mm	49 (45–53)	50 (46–57)	.022
TAPSE, cm	2.2 (2.0–2.5)	2.0 (1.6–2.2)	<.001
Moderate to severe MR, %	87 (15.7%	26 (22.8%)	.074
LAA velocity	38 (30–50)	23 (20–33)	<.001
Laboratory variables
Creatinine, mg/dl	0.86 (0.76–1.00)	0.90 (0.80–1.22)	<.001
CHA2DS2-VASc	2 (1–3)	3 (2–4)	<.001
CLOTS-AF score	3 (2–4)	4 (3–6)	<.001

AF: atrial fibrillation; TAPSE: tricuspid annular plane systolic excursion; MR: mitral regurgitation LAA: left atrial appendage.

Patients with thrombus (+) exhibited significantly larger left atrial anteroposterior (LAAP) diameters, TAPSE and LAVI, highlighting important distinctions between groups. Notably, left ventricle end-diastolic (LVED) diameter and LAA velocity showed no differences. Additionally, the mean creatinine levels were markedly elevated in the thrombus (+) cohort. The CHA₂DS₂-VASc and CLOTS-AF scores were also significantly higher in this group, further reinforcing the correlation between these scores and the presence of thrombus.

The univariable analysis and multivariable model for thrombus detection patients are summarized in Table 2. In univariate regression analysis, HT, smoking, DM, CRF, cerebrovascular accident, coronary artery disease (CAD), prior MI, paroxysmal AF, LVEF, LAVI, LVED, TAPSE, LAA velocity, creatinine, CHA₂DS₂-VASc score and CLOTS-AF score were found to predict the thrombus presence. When we entered these variables into the multivariable regression analysis, smoking CRF, paroxysmal AF, LVED, LAA velocity, CHADS₂VASc score and CLOTS-AF score were ascertained as independent determinants of thrombus presence.

Table 2.

Univariable analysis and multivariable model for thrombus presence in transesophageal echocardiography according to admission demographic and clinical characteristics, laboratory parameters, echocardiography variables.

Univariable analysis	p Value	OR (95% CI)	Multivariable analysis	p Value	OR (95% CI)
Hypertension	.002	1.194 (1.283–3.007)
Smoking	<.001	2.268 (1.470–3.499)	Smoking	<.001	2.605 (1.555–4.364)
Diabetes mellitus	.008	1.796 (1.169–2.759)
Chronic renal failure	<.001	3.061 (1.822–5.143)
Cerebrovascular accident	<.001	4.436 (1.934–10.175)
Coronary artery disease	.049	1.539 (1.002–2.363)
Prior myocardial infarction	<.001	2.461 (1.522–3.979)
Paroxysmal AF	<.001	0.231 (0.141–0.378)	Paroxysmal AF	<.001	0.253 (0.146–0.439)
Ejection fraction	<.001	0.958 (0.944–0.972)
Left atrial volume index	<.001	1.216 (1.170–1.263)
Left ventricle end-diastolic diameter	.007	1.039 (1.011–1.069)	Left ventricle end-diastolic diameter	.200	0.977 (0.942–1.013)
TAPSE	<.001	0.868 (0.827–0.911)
LAA velocity	<.001	0.918 (0.898–0.938)	LAA velocity	<.001	0.932 (0.911–0.954)
Creatinine	.002	1.783 (1.245–2.552)
CHA2DS2-VASc	<.001	1.319 (1.152–1.510)	CHADS2VASc score	.142	1.129 (0.960–1.326)
CLOTS-AF score	<.001	1.417 (1.288–1.559)	CLOTS-AF score	<.001	1.287 (1.128–1.469)

OR: odds ratio; CI: confidence interval; AF: atrial fibrillation; TAPSE: tricuspid annular plane systolic excursion; LAA: left atrial appendage.

ROC curve analysis revealed that the highest AUC curve between CHA₂DS₂-VASc and CLOTS-AF scores for the prediction of thrombus detection during TEE was observed in the CLOTS-AF score. CLOTS-AF score is ≥4 and was found to have 67% sensitivity and 65% specificity in prediction of thrombus detection (AUC: 0.70, 95% CI: 0.64–0.75, p < .001). CHA₂DS₂-VASc score is ≥3 and was found to have 62% sensitivity and 64% specificity in prediction of thrombus detection (AUC: 0.61, 95% CI: 0.55–0.67, p < .001) (Figure 2).

Figure 2.

A receiver operating characteristic (ROC) curve analyses of CLOTS-AF score and CHA₂DS₂-VASc score. AF: atrial fibrillation; AUC: area under the curve; CI: confidence interval.

4. Discussion

This study aimed to assess the efficacy of the CLOTS-AF score in identifying the presence of thrombus using TEE guidance in patients with AF. The findings revealed that the CLOTS-AF score has demonstrated an adequate level of accuracy in predicting the presence of LAAT.

Current studies suggest that most strokes in people with AF occur due to the embolization of LAAT [21]. Stroke, a serious consequence of AF, frequently leads to cardioembolic events that are not only severe but also have a significant risk of recurrence. These strokes often result in permanent disability or death [8]. The development of scores predicting LAAT is to identify high-risk patient populations and offer preventive interventions to mitigate the occurrence of severe clinical outcomes.

The CHADS₂ score, initially proposed by Gage et al. was subsequently refined and expanded into the CHA₂DS₂-VASc score by Lip et al. [20,22]. Designed primarily to assess the risk of TE and stroke, the CHA₂DS₂-VASc score exhibits significant limitations when it comes to predicting the risk of LAAT [10]. Our study corroborates previous findings, demonstrating the CHA₂DS₂-VASc score’s inadequate performance in this critical area. This raises concerns about its applicability in clinical settings where accurate risk assessment is vital for patient management. In a recent study by Segan et al., a novel scoring system named CLOTS-AF was developed, showcasing impressive predictive performance with an AUC of 0.78 for detecting LAAT in only patients undergoing TEE. Our own study corroborates these findings, revealing that the CLOTS-AF score achieved a notable predictive value of 0.70, establishing it as a significantly more reliable tool for assessing LAAT risk. Unlike our study, the Segan et al. research focused solely on patients who underwent cardioversion, and it also included those with atrial flutter, which is different from AF. Notably, the overall mean CHA₂DS₂-VASc score in their study was lower than in our analysis, suggesting they included patients at a reduced risk for TE events [11]. This study not only validated the CLOTS-AF score externally but also demonstrated its effectiveness in predicting LAAT within an entirely new cohort. By doing so, it builds upon the crucial insights provided by Segan et al. underscoring the score’s relevance and transformative potential in clinical practice.

The meta-analysis on the prevalence of LAAT in patients with AF revealed an average prevalence ranging from 2.7 to 6.3, which contrasts with our study [23]. The elevated prevalence of LAAT, recorded at 17% of the patient cohort in our study, may be linked to the population studied, where TEE was used only in patients undergoing AF ablation or cardioversion, possibly resulting in a higher incidence of LAAT. Furthermore, several studies conducted on similar cohorts have reported comparably high prevalence rates of LAAT [24,25]. Additionally, the mean CHA₂DS₂-VASc score in most studies featured in the meta-analysis was lower than that of our patients. In our findings, the CHA₂DS₂-VASc score was 2 in the thrombus (−) group and 3 in the thrombus (+) group, suggesting that our patient cohort had a significantly higher TE risk, which may contribute to the increased detection of LAAT.

In our study, we observed some notable disparities between the groups with and without LAAT, as we anticipated. These elements encompassed commonly recognized indicators such as reduced EF, higher LA anteroposterior diameter and diminished LAA emptying velocity [20,26,27]. Nevertheless, we identified elements that were absent from the CHA₂DS₂-VASc score but present in the CLOTS-AF risk score. These elements, renal impairment, persistent AF, TAPSE and LAVI, were revealed to be independent predictors.

The primary factors contributing to the high risk of TE could be the renal function levels and the type of AF, both of which are incorporated into the CLOTS-AF scoring system. In the study conducted by Bonde et al. which was considered the pioneer of the CLOTS-AF score, among patients with AF, CRF was identified as a significant risk factor for stroke and thromboembolism [28]. Not only did our study revealed that patients with CRF had a much higher prevalence of thrombus (+), but it also revealed that CRF was an independent factor in determining the existence of thrombus.

Multitude of studies has demonstrated that persistent AF is associated with a greater likelihood of stroke compared to paroxysmal AF [29]. In our study, we demonstrated that patients in the thrombus (−) group had a high rate of paroxysmal AF, while the thrombus (+) group had a high rate of persistent AF, which is consistent with previous studies.

The results of our study revealed a significant decrease in TAPSE in both the thrombus (+) group, indicating that TAPSE is a crucial independent predictor for the existence of thrombus. TAPSE is a commonly reported indicator of right ventricular (RV) systolic function and is also known to be reduced in heart failure patients with left ventricular (LV) dysfunction [30]. RV dysfunction frequently occurs in patients diagnosed with chronic obstructive pulmonary disease (COPD) and pulmonary embolism, particularly in those with LV heart failure. Pulmonary hypertension (PHT) arises due to elevated left heart filling pressures in cases of LV heart failure. These disorders can heighten the likelihood of thrombus development in the LA due to sluggish blood flow and an increased propensity to clot. Additionally, they can raise the risk of stroke by impacting cerebral blood flow [31]. Furthermore, obstructive sleep apnoea is a significant contributor to the development of PHT, and subsequent right HF and AF are frequently encountered in patients with obstructive sleep apnoea [32].

Age, HT, DM, sex and vascular disease (with the exception of a prior MI) are all factors that are included in the CHA₂DS₂-VASc score but are not included in the CLOTS-AF score. Our study did not find any evidence that these factors are independent predictors of LAAT among the participants. Given that DM and HT frequently lead to CRF and reduced EF, it is thought that the CLOTS-AF score incorporates both conditions [33,34]. While it is widely recognized that the likelihood of experiencing a stroke rises as individuals grow older, the influence of age and sex on the independent prediction of LAAT risk has not been reliably demonstrated [35–37]. The groundbreaking study by Yoshimura et al. convincingly shows that utilizing the CHA₂DS₂-VASc without the sex component found in the CHA₂DS₂-VASc score offers a simpler approach while effectively identifying TE events in patients with AF. As a result, the European Society of Cardiology (ESC) has wisely updated its guidelines, endorsing the CHA₂DS₂-VA score in place of the CHA₂DS₂-VASc score for improved patient care [38,39].

4.1. Limitations

We acknowledge that our study had several important limitations. The use of a retrospective design and the focus on a single centre could pose significant constraints. However, it was noteworthy that each patient was included consecutively, and our sample size was adequate for meaningful analysis. Furthermore, the evaluation of TEE could be quite subjective, heavily relying on the operator’s expertise, particularly in differentiating between genuine thrombus and dense SEC or sludge. This potential risk of misdiagnosing thrombus was a critical factor in our analysis. To enhance the robustness of our findings and their therapeutic implications, further prospective studies with larger sample sizes are essential.

5. Conclusions

An implementing robust scoring system is imperative to effectively predict the LAAT, which can result in critical clinical complications and significant financial burdens. This study presents compelling evidence that the CLOTS-AF score is a valuable tool in forecasting LAAT as detected TEE. Moreover, this study not only validated the CLOTS-AF score externally, but also showed its predictive power in a new patient cohort, further establishing its importance in clinical practice.

Funding Statement

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Ethical approval

This study was approved by the Ethics Committee of Health Sciences University Hamidiye Faculty (approval no. 24/638).

Consent form

Written informed consent was obtained from the patients included in this retrospective analysis.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All datasets generated during and/or analysed during the current study are not publicly available but can be provided by the corresponding author on reasonable request.