Predicting the chemical composition of urinary calculi in vivo using gray scale ultrasound

doi:10.21203/rs.3.rs-4680542/v1

Download PDF

Research Article

Predicting the chemical composition of urinary calculi in vivo using gray scale ultrasound

https://doi.org/10.21203/rs.3.rs-4680542/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objective The purpose of this study was to preliminarily predict the chemical composition of urinary calculi using ultrasound in vivo.

Methods The data of 267 urinary calculi patients were analyzed retrospectively, including non-contrast computed tomography (NCCT) and ultrasound imaging data before the intervention and post-interventional chemical composition. The Hounsfield unit (HU) value of calculi, the grayscale value of calculi and the grayscale value of posterior acoustic shadow of calculi on ultrasound were measured and analyzed statistically.

Results The chemical composition analysis indicated that there were four types of mixed calculi; the main components were calcium oxalate monohydrate (COM) calculi, calcium oxalate dihydrate (COD) calculi, carbonate apatite (CA) calculi and anhydrous uric acid (UA0) calculi. The HU value was distinguished between calcium-containing calculi and UA0 calculi, with a cut-off value of 644.00, a sensitivity of 88.00% and a specificity of 95.04%, and P < 0.001. The grayscale value of calculi on ultrasound was distinguished between calcium-containing calculi and UA0 calculi with a cut-off value of 200.29, a sensitivity of 38.84% and a specificity of 96.00%, P < 0.001. The grayscale value of the posterior acoustic shadow of calculi on ultrasound was distinguished between CA calculi and UA0 calculi with a cut-off value of 31.48, a sensitivity of 58.33% and a specificity of 84.00%, and P = 0.011.

Conclusion Ultrasound can preliminarily distinguish the chemical composition of urinary calculi and provide certain information for clinicians to choose treatment plans.

Urinary calculi

Chemical composition

NCCT

HU value

Ultrasound

Grayscale value

Urolithiasis is a common, painful and unbearable urinary system disease, which is also a risk factor for cardiovascular disease and chronic kidney disease[1]. Urolithiasis has become a major medical problem worldwide. Its high incidence and recurrence rates lead to many health problems and an increased socioeconomic burden, which has attracted more and more attention[2]. Recent epidemiological data indicate that urolithiasis affects approximately 12% of the population during their lifetime[3], and its prevalence and morbidity steadily rise worldwide[2, 4]. The prevalence of urolithiasis in the United States increased from 3.2% in 1980 to 8.8% in 2010[5]. The prevalence of urolithiasis in Asia ranged from 5–19.1%[6]. In China, due to social development, lifestyle and dietary changes, the prevalence of urolithiasis has increased from 4–6.4% in the past 30 years. In other words, 1 in 17 adults in China suffers from urolithiasis[7]. Several studies have shown that the recurrence rate of urolithiasis in cured patients increases yearly after the initial stone event[8–10]. The estimated recurrence rate is 6–17% within one year, 30%-50% within five years, and 60%-80% throughout life[11].

In the face of such a high prevalence and recurrence rate of urolithiasis, it is crucial to choose an appropriate examination method. Presently, the recognized gold standard for diagnosing urinary calculi is non-contrast computer tomography (NCCT), which is not only used for morphological and anatomical evaluation, especially the application of dual-energy computer tomography (DECT) in recent years. By measuring the inherent atomic number differences, different calculi composition shows different attenuation characteristics at different energy or kilovoltage levels. This property predicts the chemical composition of the calculi and distinguishes uric acid-containing calculi and calcium-containing calculi[12]. The advantage of DECT in identifying the calculi composition is helpful for treating patients with uric acid calculi, which can be treated by alkalizing urine to avoid surgery. At the same time, the problem of radiation dose should not be ignored because the prevalence and recurrence rate of urolithiasis continues to increase, and patients may be affected by the biological effects of ionizing radiation. Grayscale ultrasound is the preferred method for obtaining high-resolution images of non-ionizing radiation, which is economical, widely used and convenient. It can dynamically evaluate the morphology, size, quantity, anatomical information and degree of obstruction of urinary calculi in real time.

Recent research data showed that the sensitivity and specificity of gray-scale ultrasound in diagnosing urolithiasis were 45–99% and 88–94%, respectively[13–14]. A meta-analysis of a prospective study showed that the sensitivity of low-dose CT in the diagnosis of urolithiasis was 93.1% (95% confidence interval: 91.5–94.4%), and the specificity was 96.6% (95% confidence interval: 95.1–97.7%)[15]. The sensitivity and specificity of gray-scale ultrasound in diagnosing urolithiasis are no less than that of NCCT. A strong echo with posterior acoustic shadow characterizes the diagnosis of urinary calculi by gray scale ultrasound. In the process of clinical ultrasonic examination, the degree of strong echo of the calculi and the degree of echo of the posterior acoustic shadow are often different. Previous studies have shown that the calculi size, tissue medium, reverberation, distance from the probe and probe frequency may affect the degree of echo of the calculi and the posterior acoustic shadow. However, whether they are affected by chemical composition lacks a unified explanation. Therefore, the purpose of this study was to quantitatively analyze the degree of the echo of urinary calculi and posterior acoustic shadow, and attempt to explore the feasibility of gray-scale ultrasound in evaluating the chemical composition of urinary calculi in vivo.

Subjects

This study retrospectively evaluated the data of patients (age ≥ 18 years old) who underwent transurethral ureteroscopic lithotripsy or percutaneous nephrolithotomy in the Urology Department of Tianjin Medical University Second Hospital from September 2018 to January 2020. Inclusion criteria: Ultrasound examination and NCCT examination were performed before surgical treatment, and the calculi with HU value determined by NCCT were taken as the target calculi. Calculi composition results were obtained after lithotomy. Exclusion criteria: patients were treated with chemo-dissolution before the operation. The Ethics Committee of Tianjin Medical University Second Hospital approved our study design. Written informed consent from participants was not required because the study was designed retrospectively, and patient-recorded information was anonymized and de-identified before analysis.

Methods

Before the operation, all patients were examined for urinary systems by ultrasound examination and NCCT examination. Ultrasound was performed using the DC-8S ultrasound scanner (Mindry, convex array probe, 3.5 MHz) with a focus depth of 15 cm and the focus placed on the lower edge of the target calculi. The NCCT was performed using the GE Light Speed Pro 64-row helical CT (General Electrics, Health Care, Waukesha, Wisconsin, USA), with a voltage of 120 kV and a current of 240 mA. The scanning layer thickness and interval were each 5 mm, and scanning ranged from the bilateral upper pole to the pubic symphysis. All images were acquired by the picture archiving and communication systems (PACS) and stored for subsequent evaluation by two radiologists blinded to the chemical composition of the calculi during a consensus reading session.

Acquisition of HU value of target calculi: For target calculi, a region of interest (ROI) overlying the whole calculi on the slice was obtained for each plane for tissue and bone windows at 120 kV. The absolute HU value of target calculi is presented for the central ROI of the plane that passed through the maximum transverse diameter. Each target calculi was measured three times, and the average value was the HU value of the target calculi.

Measurement of related information of target calculi in gray-scale ultrasound: We used the ImageJ 1.47V (National Institute of Health, Bethesda, USA) software to circle the extent of the calculi along the boundary of the target calculi, which was the ROI, and the software automatically calculated the grayscale value. Then, the range of the posterior acoustic shadow of the target calculi was determined, and use ImageJ 1.47V software to obtain the length, width and grayscale value of the posterior acoustic shadow of the target calculi. The sonogram of each target stone is stored in at least two different sections to show the approximate shape and the extent of the calculi and to obtain a relatively objective grayscale value. The grayscale value of each target calculi and the grayscale value of the posterior acoustic shadow were measured three times and recorded the average value.

Statistical analysis

Each quantitative indicator was presented as mean ± standard deviation (mean ± SD). Continuous variables were statistically compared by one-way analysis of variance (ANOVA) and non-parametric test. After ANOVA, the least significant difference (LSD) post hoc test was used to compare groups. The Kruskal-Wallis Dunn test was used for non-parametric distributions. Receiver operating characteristics (ROC) analysis was used to calculate the cut-off HU value and grayscale value for each group to differentiate the various groups of calculi. Spearman correlation was used to test the relationship between calculi characteristics. Data were analyzed using SPSS 20.0 (SPSS, Chicago, Illinois, USA) and MedCalc 18.2 (MedCalc, Ostend, Belgium). P < 0.05 was statistically significant.

A total of 267 patients (174 Males / 93 Females) were enrolled in the study, and the mean age was 52.2 ± 12.38 years. The infrared spectroscopy chemical composition analysis showed four types of mixed composition calculi. All of the calculi of mixed composition had a primary composition calculi type contributing to at least two-thirds (> 66%) of the calculi, and the primary composition calculi also defined the group. The number of calculi in each group was as follows: calcium oxalate monohydrate calculi (COM, n = 197), calcium oxalate dihydrate calculi (COD, n = 21), anhydrous uric acid calculi (UA0, n = 25), carbonate apatite (CA, n = 24). Table 1 demonstrates the patients’ and calculi characteristics according to calculi composition.

The results showed significant differences in age, gender, calculi size on NCCT, HU value, the grayscale value of calculi on ultrasound and the length of the posterior acoustic shadow on ultrasound among the four groups (P < 0.05). There was a significant difference in the grayscale value of posterior acoustic shadow between the UA0 group and the CA group (P < 0.05). There were no significant differences in body mass index (BMI), side and location of target calculi, degree of ipsilateral hydronephrosis, calculi size from ultrasound and the width of posterior acoustic shadow on ultrasound among the four groups (P > 0.05, as shown in Table 2).

ROC analysis showed a significant difference between the UA0 group and calcium-containing calculi, including the COM group, COD group and CA group. For the HU value, the area under the curve (AUC) was 0.907, P < 0.001, and the cut-off value was 644.00 HU. For the grayscale value of calculi on ultrasound, AUC was 0.705, P < 0.001, and the cut-off value was 200.29.

Then, the results of ROC analysis of HU value, the grayscale value of calculi and the grayscale value of the posterior acoustic shadow on ultrasound were used to differentiate between calcium-containing subgroup calculi and UA0, which are depicted in Table 3.

Spearman correlation analysis was used to analyze the relationship between calculi characteristics. The results showed that the calculi size on ultrasound was positively correlated with the width of posterior acoustic shadow on ultrasound, and the correlation coefficient was 0.755, P < 0.001. The HU value was positively correlated with the grayscale value of calculi on ultrasound, and the correlation coefficient was 0.121, P < 0.05.

The high prevalence and recurrence rate of urolithiasis has attracted more and more attention. A complete understanding of calculi information before treatment helps select a reasonable and practical treatment plan and improve the treatment success rate, such as the composition type, size and location of calculi. The composition of calculi has different physical and solid properties, which directly affect the success of treatment[16]. Calcium oxalate calculi are the most common composition of urolithiasis. It has a dense structure, hard texture, smooth surface or mulberry shape, and the treatment effect of extracorporeal shock wave lithotripsy (ESWL) is poor. Carbonate apatite is an infectious calculi which should be removed as much as possible during the operation, and urinary tract infection and acidified urine should be actively controlled after operation to prevent recurrence and rapid growth of calculi. Uric acid calculi are soft and can be treated with medicine. Therefore, understanding the composition of calculi before treatment is both an advantage and a trend.

Currently, there are many studies on applying NCCT to predict the composition of urinary calculi. Torricelli et al. found that the HU value of cystine calculi was 648 ± 122 HU and that of calcium oxalate calculi was 1099 ± 239 HU, which successfully distinguished calcium oxalate calculi from cystine calculi[17]. Pareek et al. found that HU value ≤ 500 HU and urinary pH ≤ 5.5 could be used to distinguish uric acid calculi from calcium-containing calculi[18–19]. Similarly, the HU value of calcium-containing calculi was higher in adults compared with uric acid calculi and cystine calculi[17]. Lee et al. showed that the HU value of uric acid calculi, phosphate calculi and calcium oxalate calculi were 513 ± 197 HU, 1660 ± 292 HU and 1684 ± 290 HU, respectively, which successfully distinguishing uric acid calculi from calcium-containing calculi[20]. In our study, we distinguished the composition of the subgroup calculi, and the results showed that the cut-off value of calcium oxalate dihydrate calculi for distinguishing carbonate apatite calculi and anhydrous uric acid calculi were 1063.00HU and 644.00 HU, respectively. The cut-off value of anhydrous uric acid calculi to distinguish carbonate apatite and calcium oxalate monohydrate calculi was 644.00 HU, respectively. In summary, an HU value ≤ 644.00 could be used to distinguish calcium-containing calculi from anhydrous uric acid calculi. Our results are consistent with previous studies, showing that the HU value of calcium-containing calculi is higher than that of anhydrous uric acid calculi, and the HU value of NCCT can be used to distinguish calcium-containing calculi from anhydrous uric acid calculi.

Although NCCT has become the gold standard for diagnosing of urolithiasis, ultrasound is widely used to diagnose urinary calculi in the daily examination, which is economical, convenient and safe. Recent studies have reported that the sensitivity and specificity of ultrasound in the diagnosis of urolithiasis are 90–93% and 95–100%, respectively[21]. Ultrasound can provide clinicians with rich information, including the calculi size, number, location, and degree of obstruction. The accuracy and information ultrasound provides in diagnosing urinary calculi are similar to that of NCCT. Therefore, in this study, we made a quantitative analysis of the strong echo and posterior acoustic shadow by using the ultrasonic characteristics of urinary calculi to analyze the feasibility of the preliminary prediction of the composition of urinary calculi by ultrasound. The results showed that the grayscale value of strong echo of calcium-containing calculi in this study was higher than that of anhydrous uric acid calculi. The cut-off value of calcium-containing calculi and anhydrous uric acid calculi was 200.29; that is, the grayscale value of strong echo of calculi on ultrasound was ≤ 200.29, which can preliminarily distinguish calcium-containing calculi and anhydrous uric acid calculi. The results were similar to those of NCCT in predicting calculi composition.

For the posterior acoustic shadow of urinary calculi, the degree of the ultrasonic manifestation is different, and the influence of the different chemical compositions of urinary calculi on the posterior acoustic shadow has been widely discussed. King et al. showed that the chemical composition of the calculi had no apparent effect on the posterior acoustic shadow. At the same time, the rough texture of the tissue could affect the posterior acoustic shadow behind the small calculi[22]. Rubin et al. revealed that surface topography, such as roughness and radian of calculi, could affect the clarity of the posterior acoustic shadow[23]. The posterior acoustic shadow of the calculi with a rough surface and small radian is clear. In contrast, the posterior acoustic shadow of calculi with a smooth surface and large radian is shallow. However, some scholars indicated that the more calcium in tissues, the more pronounced the sound attenuation[24], i.e., the lower the degree of posterior acoustic shadow echo and the lower the grayscale value. According to our findings, the posterior acoustic shadow grayscale values in the CA group were lower than in the UA0 group and more apparent than in the UA0 group, which indicates a relationship between the chemical composition of the calculi and the echo degree of the posterior acoustic shadow, which may be related to the difference in the number of calcium elements and elements of different substances. The CA group contains high calcium and has more substantial ultrasonic acoustic attenuation, lower echo, and more apparent posterior acoustic shadow. In contrast, the UA0 group does not contain calcium elements, has relatively weaker acoustic attenuation, higher echo, and produces lighter acoustic shadow and higher grayscale value. The results of calculi composition analysis by using energy spectral CT showed that high atomic number substances mainly produce photoelectric absorption effects. Carbonate apatite contains high-order chemical elements (Ca, P, S) with strong x-ray attenuation. In contrast uric acid contains low-order chemical elements (H, C, N, O) with low x-ray attenuation, suggesting a correlation between acoustic attenuation and the order of chemical elements. This is similar to our findings that calcium-containing calculi are significantly more posteriorly attenuated than non-calcium-containing calculi, with lower posterior acoustic echoes and grayscale values.

This study also found a relationship between the HU value of calculi and the grayscale value of calculi strong echo. It is suggested that the echo of calculi on ultrasound is more substantial, its HU value may be more significant, and its calcium content is more. The more pronounced the posterior attenuation is, the lower the echo of the posterior acoustic shadow is. On the contrary, the echo of calculi on ultrasound is reduced. Its HU value may be smaller, with less calcium or non-calcium content, and the posterior acoustic shadow is lighter. The results of this study still have some implications for the treatment of urolithiasis. For example, uric acid calculi, negative calculi without X-ray development, can be identified by ultrasound. The grayscale value of the strong echo and the posterior acoustic shadow determined the calculi properties. Uric acid calculi can be dissolved clinically by alkalinizing urine with drugs to avoid ESWL or surgical treatment[25]. In addition, the treatment of urolithiasis is not only concerned with the chemical composition of the calculi; calculi size is also a critical factor in determining the treatment plan for urolithiasis[26]. In order to improve the accurate measurement of the calculi size by ultrasound, the width of the posterior acoustic shadow is used as an additional measure, especially for calculi size ≤ 5 mm[27–29]. Most calculi without posterior acoustic shadows were ≤ 5 mm, and these had the highest possibility of spontaneous calculi discharge[30]. Therefore, the accurate measurement of calculi size by posterior acoustic shadow is also of guiding significance for urolithiasis treatment.

In this study, the sensitivity and specificity of using HU value to distinguish calcium-containing calculi from anhydrous uric acid calculi were 85.87% − 95.24% and 66.67% − 95.43%, respectively. The sensitivity and specificity were 38.84%-96.00% and 37.56% − 96%, respectively, for distinguishing calcium-containing calculi from anhydrous uric acid calculi using the grayscale value of the calculi and the posterior acoustic shadow on ultrasound. Although the sensitivity of ultrasound to predict urinary calculi composition is lower than that of NCCT, it may be due to the small number of patients in the subgroup of this study. In addition, the ROIs of the strong echo of calculi and the posterior acoustic shadow are artificially determined. Determining the possible boundary may need to be more accurate to affect the study results. However, this study quantified the characteristics of calculi on ultrasound. It made a preliminary prediction of calculi composition, which provided an empirical basis for future research, aiming to hope that ultrasound may be a valuable tool for predicting urinary calculi composition and providing more information for the treatment choice and follow-up review of urolithiasis.

In this study, calcium-containing calculi and anhydrous uric acid calculi were distinguished by measuring the grayscale value of the strong echo and the posterior acoustic shadow of urinary calculi on ultrasound. Ultrasound can provide some information for predicting the chemical composition of urinary calculi and help clinicians choose the appropriate treatment plans.

Acknowledgements

We extend our sincere gratitude to our department chair for the support. We also thank our physicians, engineers, nurses, and other department staff.

Authors^,contributions

LJ designed the study, LNN collected data and wrote the manuscript, and LJ revised the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by the Second Hospital of Tianjin Medical University Youth Research Fund Project (2023ydey16).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no competing interests.

Ethical approval and Informed consent

Ethical approval was obtained by the local institutional review board (Ethical Committee of Second Hospital of Tianjin Medical University). Furthermore, written informed consent from participants was not required because the study was designed retrospectively, and patient-recorded information was anonymized and de-identified before analysis.

Availability of data and materials

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

Cheungpasitporn W, Thongprayoon C, Mao MA et al (2014) The Risk of Coronary Heart Disease in Patients with Kidney Stones: A Systematic Review and Meta-analysis. N Am J Med Sci 6(11):580-585
Kittanamongkolchai W, Vaughan LE, Enders FT et al (2018) The Changing Incidence and Presentation of Urinary Stones Over 3 Decades. Mayo Clin Proc 93(3):291-299
Alelign T, Petros B (2018) Kidney stone disease: An update on current concepts. Adv Urol 2018:3068365
Wang P, Zhang H, Zhou J et al (2021) Study of risk factor of urinary calculi according to the association between stone composition with urine component. Sci Rep 11(1):8723
Scales CD Jr, Smith AC, Hanley JM et al (2012) Prevalence of kidney stones in the United States. Eur Urol 62(1):160-165
Sakamoto S, Miyazawa K, Yasui T et al (2019) Chronological changes in epidemiological characteristics of lower urinary tract urolithiasis in Japan. Int J Urol 26(1):96-101
Zeng G, Mai Z, Xia S et al (2017) Prevalence of kidney stones in China:An ultrasonography based cross-sectional study. BJU Int 120:109–116
Daudon M, Jungers P, Bazin D et al (2018) Recurrence rates of urinary calculi according to stone composition and morphology. Urolithiasis 46:459–470
D'Costa MR, Haley WE, Mara KC et al (2019) Symptomatic and Radiographic Manifestations of Kidney Stone Recurrence and Their Prediction by Risk Factors: A Prospective Cohort Study. J Am Soc Nephrol 30(7):1251-1260
Ziemba JB, Matlaga BR (2017) Epidemiology and economics of nephrolithiasis. Investig Clin Urol 58:299–306
Huang WY, Chen YF, Carter S et al (2013) Epidemiology of upper urinary tract stone disease in a Taiwanese population: a nation wide, population based study. J Urol 189(6):2158–2163
Mahalingam H, Lal A, Mandal AK et al (2015) Evaluation of low-dose dual energy computed tomography for in vivo assessment of renal/ureteric calculus composition. Korean J Urol 56(8):587-593
Ray AA, Ghiculete D, Pace KT et al (2010) Limitations to ultrasound in the detection and measurement of urinary tract calculi. Urology 76(2):295-300
Smith-Bindman R, Aubin C, Bailitz J et al (2014) Ultrasonography versus computed tomography for suspected nephrolithiasis. N Engl J Med 371(12):1100-1110.
Xiang H, Chan M, Brown V et al (2017) Systematic review and meta-analysis of the diagnostic accuracy of low-dose computed tomography of the kidneys, ureters and bladder for urolithiasis. J Med Imaging Radiat Oncol 61(5):582-590
Altan M, Çitamak B, Bozaci AC et al (2017) Predicting the stone composition of children preoperatively by Hounsfield unit detection on non-contrast computed tomography. J Pediatr Urol 13(5):505.e1-505.e6
Torricelli FC, Marchini GS, De S et al (2014) Predicting urinary stone composition based on single-energy noncontrast computed tomography: the challenge of cystine. Urology 83:1258e63
Pareek G, Armenakas NA, Fracchia JA (2003) Hounsfifield units on computerized tomography predict stone-free rates after extracorporeal shock wave lithotripsy. J Urol 169:1679e81.
Spettel S, Shah P, Sekhar K et al (2013) Using Hounsfifield unit measurement and urine parameters to predict uric acid stones. Urology 82:22e6.
Lee JS, Cho KS, Lee SH et al (2018) Stone heterogeneity index on single-energy noncontrast computed tomography can be a positive predictor of urinary stone composition. PLoS One 13(4):e0193945
Abdel-Gawad M, Kadasne RD, Elsobky E et al (2016) A prospective comparative study of color doppler ultrasound with twinkling and noncontrast computerized tomography for the evaluation of acute renal colic. J Urol 196(3):757–762
King W 3rd, Kimme-Smith C, Winter J (1985) Renal stone shadowing: an investigation of contributing factors. Radiology 154(1):191-196
Rubin JM, Adler RS, Bude RO et al (1991) Clean and dirty shadowing at US: a reappraisal. Radiology 181(1):231-236
Zhou YC, Guo WX. Ultrasound medicine. 6thed. Beijing:People's Military Medical Publishing House. 2013.25.
Kenny JE, Goldfarb DS (2010) Update on the pathophysiology and management of uric acid renal stones. Curr Rheumatol Rep 12(2):125-129
Hubner WA, Irby P, Stoller ML (1993)Natural history and current concepts for the treatment of small ureteral calculi. Eur Urol 24:172–176.
Dunmire B, Harper JD, Cunitz BW et al (2016) Use of the Acoustic Shadow Width to Determine Kidney Stone Size with Ultrasound. J Urol 195(1):171-177
May PC, Haider Y, Dunmire B et al (2016) Stone-Mode Ultrasound for Determining Renal Stone Size. J Endourol 30(9):958-962
Dunmire B, Lee FC, Hsi RS et al (2015) Tools to improve the accuracy of kidney stone sizing with ultrasound. J Endourol 29(2):147-152
Dai JC, Dunmire B, Sternberg KM et al (2018) Retrospective comparison of measured stone size and posterior acoustic shadow width in clinical ultrasound images. World J Urol 36(5):727-732

Tables 1-3 are available in the Supplementary Files section.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Predicting the chemical composition of urinary calculi in vivo using gray scale ultrasound

Status:

Version 1

Abstract

Introduction

Materials and Methods

Subjects

Methods

Statistical analysis

Results

Discussion

Conclusion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1