Characteristics and management of mechanically ventilated patients in South Korea compared with other high-income Asian countries and regions
Article information
, Kyeongman Jeon2, Suk-Kyung Hong3, Ah Young Leem4, Jee Hwan Ahn5, Hang Jea Jang6, Ki Sup Byun7, So Hee Park5, Sojung Park8, Yoon Mi Shin9, Jisoo Park10, Sung Wook Kang11, Jin Hyoung Kim12, Jinkyeong Park11, Deokkyu Kim13, Bo Young Lee14, Woo Hyun Cho15, Kwangha Lee16, Song I Lee17, Tai Sun Park18, Yun Jung Jung19, Sang-Hyun Kwak20, Sang-Beom Jeon21, Sung Hyun Kim22, Won Jai Jung23, Sang-Min Lee24, Sunghoon Park25, Yun Su Sim26, Young-Jae Cho27, Younsuck Koh,5
Abstract
Background
This study investigated the characteristics of mechanically ventilated patients in South Korean intensive care units (ICUs).
Methods
We conducted a subgroup analysis of a multinational observational study. Data from 271 mechanically ventilated patients in South Korean ICUs were analyzed for demographics, ventilation practices, and mortality, and were compared with those of 327 patients from other high-income Asian countries.
Results
South Korean patients were older (mean age: 67 vs. 62 years, P<0.001) and had lower ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen (255.5 vs. 306.2, P<0.001). South Korean ICUs exhibited higher patient-to-nurse ratios (2.6 vs. 1.9, P<0.001) and more beds per unit (20.5 vs. 16.0, P=0.017). The use of sufficient positive end-expiratory pressure for patients (PEEP) for acute respiratory distress syndrome (ARDS) was less frequent in South Korea (62.2% vs. 91.2%, P=0.005). Mortality rates were similar between South Korean patients and those in other high-income Asian countries (38.0% vs. 34.2%, P=0.401). Significant mortality predictors in South Korea included age ≥65 years (odds ratio [OR], 4.03; P=0.039) and a Sequential Organ Failure Assessment score ≥8 (OR, 2.36; P=0.031). The presence of respiratory therapists was associated with reduced mortality (OR, 0.52; P=0.034).
Conclusions
Despite higher age and patient-to-nurse ratios in South Korean ICUs, outcomes were comparable to those in other high-income Asian countries. The suboptimal use of sufficient PEEP with ARDS indicates potential areas for improvement. Additionally, the beneficial impact of respiratory therapists on mortality rates warrants further investigation.
INTRODUCTION
Mechanical ventilation is a cornerstone of intensive care unit (ICU) management, crucial for reducing the labor of breathing and ensuring effective gas exchange in critically ill patients [1,2]. Despite its therapeutic benefits, mechanical ventilation must be approached with caution to prevent potential iatrogenic lung injuries [3-5]. Implementing lung-protective strategies, such as appropriate tidal volume settings and adjunctive therapies, is essential for mitigating risks and enhancing patient outcomes [6-9]. Optimal ventilatory management requires vigilant monitoring and timely interventions, which underscore the importance of well-resourced and expertly staffed ICUs [10-14].
To examine the current state of mechanical ventilation practices across Asian ICUs, the Asian Critical Care Clinical Trials (ACCCT) group undertook a comprehensive multinational observational study [15]. Encompassing 190 ICUs across 19 Asian countries, the study enrolled 1,408 patients undergoing invasive mechanical ventilation and revealed noteworthy insights into regional practices and challenges. The results reveal that low-tidal volume ventilation and adequate positive end-expiratory pressure (PEEP) were underused in patients with acute respiratory distress syndrome (ARDS) in Asia. Furthermore, the study found that country income, patient age, and illness severity were significant factors influencing the mortality rates of mechanically ventilated patients.
South Korea (hereafter, Korea), classified as a high-income country by the World Bank Group, had an average ICU bed availability of 17.6 ICU beds per 100,000 population in 2022, placing it in the upper-middle range among high-income Asian countries [16]. Despite that infrastructure, prior research has highlighted a significant shortage of intensivists in Korean ICUs. As of 2014, only 40 ICUs (29.4%) employed ICU specialists working a 5-day week [17]. A nationwide registry data study further revealed that merely 42.4% of patients were admitted to ICUs staffed by trained intensivists, with 57.8% of patients treated in units lacking such expertise [18].
Given those challenges, we conducted a subgroup analysis of the ACCCT group study [15] to better understand mechanical ventilation practices within Korean ICUs. Although practices can vary significantly among individual ICUs, prior studies have shown that ICU infrastructure is closely associated with a country's income level [19]. Therefore, we used data from the original ACCCT study to compare the mechanical ventilation practices in Korean ICUs with those in other high-income Asian countries. Our objectives were to describe patient demographics, evaluate current ventilation practices, and explore factors that influence mortality among mechanically ventilated patients in Korean ICUs, with a focus on identifying differences and similarities with other high-income Asian countries.
MATERIALS AND METHODS
This study was conducted in accordance with the principles of the Declaration of Helsinki. The Institutional Review Board of each participating hospital approved patient enrollment; primary approval was granted by the Institutional Review Board of Asan Medical Center (No. 2018-0865). It involved a secondary analysis of previously collected, de-identified data; therefore, the requirement for patient informed consent was waived.
Study Design and Data Collection
We accessed data from a multicenter cross-sectional study conducted by the ACCCT group, which included 91 ICUs from the high-income Asian countries of Brunei, Japan, Saudi Arabia, Singapore, Korea, and Taiwan. To analyze ICU characteristics and staffing, we compared Korean facilities with those from other high-income Asian countries and regions. Specifically, we compared the type of hospital (public vs. private, teaching vs. non-teaching), type of ICU (general ICU, medical ICU, surgical ICU, and others), ICU practice model (open, co-management, and closed), number of beds per ICU, patient-to-nurse ratio, presence of a designated critical care attending physician, and availability of a respiratory therapist.
Using patient data from the ACCCT group study, Korean patients were compared with a pooled dataset of patients from other high-income Asian countries and regions (Brunei, Japan, Saudi Arabia, Singapore, and Taiwan). Subjects for whom the reason for mechanical ventilation was postoperation care were excluded from this analysis (Figure 1). We examined demographics, disease severity, reasons for mechanical ventilation, adjunctive treatments, mechanical ventilation practices, use of lung-protective strategies, and clinical outcomes.
Flowchart of study population selection.
The original ACCCT study enrolled patients older than 18 years who were undergoing invasive mechanical ventilation on the day of data collection. Patient characteristics and mechanical ventilation data were gathered at 10 AM on November 4, 2019, and clinical outcomes were tracked 28 days later, with final data collection at 10 AM on December 3, 2019. The detailed data collection protocol is described in the original study [15].
Suitable PEEP levels for ARDS patients are defined in the original ACCCT study, which adopted the ARDSNet study protocol [6]: PEEP ≥5 cm H2O for FiO2, 0.21–0.49; PEEP ≥8 cm H2O for FiO2, 0.50–0.59; PEEP ≥10 cm H2O for FiO2, 0.60–0.79; PEEP ≥14 cm H2O for FiO2, 0.80–0.99; and PEEP ≥18 cm H2O for FiO2, 1.0. In the ACCCT study, patients recorded with ARDS as the reason for mechanical ventilation and patients who met the Berlin criteria [20] were classified as having ARDS.
Statistical Analysis
We analyzed the data using proportions (percentages) for categorical variables and means±standard deviations for continuous variables. Fisher’s exact test was used to compare categorical variables between groups, and the t-test was used to assess significant differences in continuous variables. To compare mortality rates, propensity score matching was used with the following variables: age, sex, body mass index (BMI), Sequential Organ Failure Assessment (SOFA) score, the ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen (P/F ratio), ARDS, use of steroid or neuromuscular blocker, prone position, number of vasopressors used, patients enrolled from teaching hospital, and type of ICU (general, medical, surgical, and others).
To identify predictors of 28-day mortality, we conducted both univariable and multivariable logistic regression analyses. The potential factors were patient age; sex; BMI; SOFA scores; presence of ARDS; P/F ratio; use of low-tidal volume ventilation; limited plateau pressure; and use of vasopressors, steroids, neuromuscular blockers, analgesics, sedatives, antipsychotics, and bronchodilators. Additionally, we included hospital type (public vs. private), ≥20 ICU beds, patient-to-nurse ratio <2, closed ICU, availability of respiratory therapists, and presence of a designated critical care attending physician.
The multivariable logistic regression included variables with P-values <0.1 in the univariable analysis. Patients with missing outcome data were excluded from these analyses. For all the analyses, statistical significance was set at a P-value of 0.05. All analyses except propensity score matching were conducted using Stata version 16.1 (StataCorp.). The propensity score matching was performed with R software version 4.4.1 (R Foundation for Statistical Computing, 2024).
RESULTS
Demographic Characteristics
We enrolled 271 mechanically ventilated patients from 31 ICUs in Korea. As detailed in Table 1, the mean age of these patients was 67 years, and 62.0% were male. The mean P/F ratio was 255.5, and 13.7% of the patients were classified as having ARDS according to our criteria. The most prevalent reason for mechanical ventilation was pneumonia (40.6%), followed by respiratory distress (33.6%) and hypoxic respiratory failure (21.0%). As adjunctive treatments, analgesics were administered to 57.8% of patients, and 47.2% received sedatives. Furthermore, 35.4% of patients received vasopressors, 26.3% were given steroids, 6.6% received neuromuscular blockers, and 0.4% were in the prone position. Additionally, 14.1% were given antipsychotics, and 31.0% were treated with bronchodilators.
Demographic characteristics, reasons for mechanical ventilation, and adjunctive treatments of mechanically ventilated patients in South Korean intensive care units compared to other high-income Asian countries
Compared with 327 patients from 60 ICUs in the other high-income Asian countries, Korean patients were significantly older (66.6 years vs. 61.7 years, P<0.001) and had a significantly lower BMI (22.3 vs. 24.9, P<0.001), a lower mean SOFA score (7.44 vs. 8.47, P=0.008), and a lower mean P/F ratio (255.5 vs. 306.2, P<0.001). Higher proportions of Korean patients received antipsychotics (14.1% vs. 7.3%, P=0.010) and bronchodilators (31.0% vs. 11.6%, P<0.001).
Mortality Rate
The mortality rates before and after propensity matching are described in Table 2. The 28-day mortality rate before propensity matching for patients in Korean ICUs was 38.0%. No significant differences in mortality rates were observed between Korean patients and those from other high-income Asian countries (38.0% vs. 34.2%, P=0.401). After propensity score matching, the mortality rates remained comparable between the two groups (36.7% vs. 40.0%, P=0.759).
Mortality rates with propensity score matching between mechanically ventilated patients in South Korean intensive care units and those in other high-income Asian countries
Mechanical Ventilation Practices and Lung-Protective Ventilation
Table 3 provides a comprehensive overview of mechanical ventilation parameters and lung-protective ventilation practices in Korean ICUs and compares them with those in other high-income Asian countries. In Korea, pressure control was the most frequently used ventilation mode (47.3%), followed by pressure support (20.7%) and synchronized intermittent mandatory ventilation (pressure control) (19.7%). In other high-income Asian countries, pressure control (28.7%) was the most common mode, followed by pressure support (28.0%) and volume control (26.6%).
Mechanical ventilation parameters and lung-protective ventilation compliance rate
The mean tidal volume for mechanically ventilated patients in Korean ICUs was 7.17 ml/kg predicted body weight (PBW), significantly lower than the 7.86 ml/kg PBW observed in other high-income Asian countries (P<0.001). The mean plateau pressure was 19.8 cm H2O in Korea, which was not significantly different from the 19.6 cm H2O recorded in other high-income countries (P=0.713). The mean PEEP was 6.43 cm H2O in Korea, with no significant difference from the 6.54 cm H2O recorded in other high-income Asian countries (P=0.504).
For patients with ARDS, Korean ICUs used a mean tidal volume of 6.96 ml/kg PBW, comparable to the 7.37 ml/kg PBW used in other high-income Asian countries (P=0.496). The mean plateau pressure and PEEP values for ARDS patients in Korean ICUs were 24.00 cm H2O and 7.49 cm H2O, respectively, with no significant differences from the 23.18 cm H2O and 8.34 cm H2O in other high-income Asian countries (plateau pressure, P=0.635; PEEP, P=0.163). Among ARDS patients, 73.0% in Korea were on low-tidal volume ventilation (tidal volume, ≤8 ml/kg PBW), and 83.7% were on limited plateau pressure (plateau pressure <30 cm H2O). The use of low tidal volume and limited plateau pressure was comparable to other high-income Asian countries. However, only 62.2% of Korean ICU patients were on sufficient PEEP, which was significantly lower than the 91.2% reported in other high-income Asian countries (P=0.005).
For non-ARDS patients, the mean tidal volume in Korean ICUs was 7.16 ml/kg PBW, which was lower than the 7.81 ml/kg PBW in other high-income countries (P=0.002). The mean plateau pressure and PEEP values were 19.52 cm H2O and 6.33 cm H2O, respectively, and were not significantly different from the 19.26 cm H2O and 6.28 cm H2O in other high-income Asian countries (plateau pressure, P=0.663; PEEP, P=0.767). A higher proportion of non-ARDS patients in Korea was on intermediate tidal volume ventilation (tidal volume, ≤10 ml/kg PBW) (89.3% vs. 83.3%, P=0.016) and limited plateau pressure (plateau pressure, <30 cm H2O) (86.3% vs. 76.1%, P=0.004).
Staffing and Infrastructure in ICUs
The organizational characteristics, practice models, staffing, and ancillary services of the 31 ICUs from Korea and 60 ICUs from other high-income Asian countries and regions are described in Table 4. In Korea, all participating ICUs were based in teaching hospitals, with a notable predominance of private institutions (67.7%) compared to public ones (32.3%). This contrasts sharply with the other high-income Asian countries, where public teaching hospitals were more common (73.3%), and private teaching hospitals were significantly fewer (11.7%). In Korean ICUs, the co-management model involving a consultant intensivist was most common (51.6%), whereas the closed ICU model, in which a full-time intensivist oversees most patient care, was the most common model in other high-income Asian ICUs (73.3%).
Organization characteristics, practice models, staffing, and ancillary services of intensive care units in South Korea compared to other high-income Asian countries
Regarding infrastructure, Korean ICUs had a significantly higher average number of beds per unit (20.5 beds) than ICUs in other high-income Asian countries (16.0 beds) (P=0.017). The number of patients designated to each nurse (patient-to-nurse ratio) was also higher in Korea (2.6) than in other high-income Asian countries (1.9) (P<0.001). Notably, fewer Korean ICUs had patient-to-nurse ratios of 1:1 and 2:1 than ICUs in other high-income Asian countries (3.2% vs. 25.0% and 48.4% vs. 70.0%, respectively). Meanwhile, more Korean ICUs had a patient-to-nurse ratio of 3:1 (45.2% vs. 1.7%).
The proportion of ICUs with designated critical care attending physicians was similar between Korea (87.1%) and other high-income Asian countries (91.7%) (P=0.484). Although fewer Korean ICUs had respiratory therapists (41.9%) than those in other high-income Asian countries (60.0%), this difference was not significant (P=0.123).
Factors Associated with Mortality
Table 5 shows the results of univariable and multivariable logistic regression analyses identifying predictors of 28-day mortality. In Korean ICUs, the univariable logistic regression analysis identified age, specifically age 65 years or older, as a significant predictor of 28-day mortality (OR, 4.24; 95% CI, 1.17–15.41; P=0.028). In the multivariable logistic regression analysis, the following covariates were included based on P-value <0.1 in univariable analysis: age, SOFA score, antipsychotic use, and respiratory therapist. The multivariable analysis confirmed that age ≥65 years (OR, 4.03; 95% CI, 1.07–15.14; P=0.039), SOFA score 8–11 (OR, 2.36; 95% CI, 1.08–5.14; P=0.031), and SOFA score 12–24 (OR, 2.47; 95% CI, 1.02–6.00; P=0.045) were significantly associated with increased 28-day mortality. The presence of a respiratory therapist was associated with decreased 28-day mortality (OR, 0.52; 95% CI, 0.28–0.95; P=0.034). The multivariable model for Korean ICUs had a P-value of 0.002 and an area under the receiver operating characteristic curve (AUROC) of 0.678.
Predictors of 28-day mortality with univariable and multivariable logistic analysis of the mechanically ventilated patients in South Korea and other high-income Asian countries
Among patients in other high-income Asian ICUs, univariable analysis found significant associations between 28-day mortality and age 41–64 (OR, 3.01; 95% CI, 1.07–8.45; P=0.036), age ≥ 65 (OR, 4.44; 95% CI, 1.64–12.07; P=0.003), SOFA score 8–11 (OR, 3.36; 95% CI, 1.51–7.47; P=0.003), SOFA score 12–24 (OR, 5.57; 95% CI, 2.52–12.32; P<0.001), P/F ratio 100–199 (OR, 2.88; 95% CI, 1.30–6.40; P=0.009), plateau pressure <30 cm H2O (OR, 0.55; 95% CI, 0.31–0.99; P=0.049), and use of steroids (OR, 2.62; 95% CI, 1.52–4.52; P=0.001) and bronchodilators (OR, 2.27; 95% CI, 1.07–4.81; P=0.033). In the multivariable logistic regression analysis, the following covariates were included based on P-value <0.1 in univariable analysis: age; SOFA score; P/F ratio; plateau pressure <30 cm H2O; use of vasopressors, steroids, or bronchodilators; and patient-to-nurse ratio <2. The multivariable analysis revealed that age 41–64 (OR, 5.04; 95% CI, 1.51–16.82; P=0.009), age ≥65 (OR, 8.16; 95% CI, 2.49–26.73; P=0.001), SOFA score 8–11 (OR, 2.73; 95% CI, 1.13–6.62; P=0.026), SOFA score 12–24 (OR, 5.10; 95% CI, 2.00–12.97; P=0.001), and use of steroids (OR, 2.82; 95% CI, 1.50–5.28; P=0.001) were significantly associated with increased 28-day mortality. The P-value for this multivariable model was <0.001, with an AUROC of 0.754.
DISCUSSION
This subgroup analysis examines the characteristics of mechanically ventilated patients and ventilator practices in Korean ICUs. Compared with patients in other high-income Asian countries, Korean patients were significantly older and had a lower mean P/F ratio. However, they also showed a lower mean SOFA score. Despite those differences, 28-day mortality did not differ significantly between Korea and other high-income Asian countries. In terms of ventilation practices, Korean ICUs applied lower PEEP levels and showed weaker adherence to sufficient PEEP in ARDS patients, though the use of low tidal volumes was similar. Regarding infrastructure, Korean ICUs had a significantly higher average number of beds per unit than ICUs in other high-income Asian countries, and the patient-to-nurse ratio was also higher in Korea. Age ≥65 years and SOFA score ≥8 were associated with increased 28-day mortality. Notably, the presence of a respiratory therapist in Korean ICUs was associated with decreased 28-day mortality.
One prominent demographic observation in our data is the significantly higher proportion of elderly patients in Korean ICUs. We found that 23.6% of mechanically ventilated patients in Korean ICUs were aged 80 or older, nearly double the rate observed in other high-income Asian countries (13.2%, P=0.001). This higher prevalence of very old patients can be attributed to cultural norms in Korea, where end-of-life decisions are frequently made by families rather than by patients themselves [21-23]. The familial inclination toward "full treatment," which includes intubation and mechanical ventilation, can lead to increased use of these interventions among older patients. This cultural context likely contributes to the higher rates of mechanical ventilation in the elderly population.
The infrastructure and staffing of Korean ICUs also present noteworthy challenges. Our data indicate that Korean ICUs have a higher number of beds per unit and a greater patient-to-nurse ratio than ICUs in other high-income Asian countries. In Korea, the scarcity of intensivists typically results in each unit being supported by a single designated intensivist. According to a 2020 government report, the average number of ICU beds per intensivist in Korea was 22.2 [24]. Our study yielded similar results, with a mean of 20.5 beds per ICU in Korea, significantly higher than the average of 16.0 in other high-income Asian countries. Additionally, the number of designated patients per ICU nurse was significantly higher in Korea, with 2.6 designated patients per ICU nurse compared to 1.9 in other high-income Asian countries.
Despite the older patient demographic and possibly higher workload faced by ICU staff in Korea, the 28-day mortality rate remained comparable to that in other high-income Asian countries. Similar to those countries, age and illness severity were key factors associated with mortality in Korean ICUs. Interestingly, our analysis shows that the presence of a respiratory therapist in a Korean ICU significantly reduced 28-day mortality. However, because the role of respiratory therapists is not standardized in Korea, it is challenging to attribute this benefit solely to their presence.
Upon reviewing the dataset to understand this finding, we found that hospitals employing respiratory therapists tended to be larger facilities and had the highest bed capacities in Korea. These hospitals likely have superior human resources, including more intensivists, trainee physicians, nurses, and ancillary staff, as well as advanced operating systems such as rapid response teams [25] and 24-hour interventional radiology services. Unfortunately, those variables were not captured in our dataset, representing a limitation in our ability to explain the observed mortality benefit. Further research is needed to identify the factors that contribute to the improved clinical outcomes associated with the presence of a respiratory therapist in Korean ICUs.
We observed no significant association between the presence of a designated attending physician in the ICU and patient mortality. Although many studies have shown that intensivists are associated with improved outcomes in critically ill patients [26-28], some studies have reported differing results [29-33]. Factors such as ICU characteristics, number and expertise of intensivists, multidisciplinary team rounds, and broader structural variations in facility and healthcare systems likely influence these outcomes. In our study, detailed data on intensivist qualifications and roles were not collected, potentially introducing residual confounding that could affect our results. The imbalance in the enrolled ICUs, with most having designated critical care medicine attendants (87.1% of Korean ICUs and 91.7% in other high-income Asian countries) and most patients coming from those ICUs (91.4% of Korean patients and 96.0% in other high-income Asian countries) might also have contributed to our inability to effectively capture the effects of intensivists.
In terms of mechanical ventilation practices, our analyses show that Korean ICUs do not differ from other high-income Asian countries in their use of low-tidal volume ventilation and limited plateau pressure for ARDS patients. However, less frequent use of sufficient PEEP is a significant difference in Korea, although the mortality rate for ARDS patients did not differ significantly from that in the other countries (37.1% vs. 35.7%, P=1.000) (Table 3).
As adjunctive treatments, more patients in Korea received bronchodilators than in other high-income Asian countries. That might suggest that mechanically ventilated patients in Korea have more frequent underlying reactive airway diseases, or bronchodilators might be routinely administered to mechanically ventilated patients [34]. Another significant finding is that more patients received antipsychotics in Korea than in other high-income Asian countries. However, the use of sedatives did not differ significantly [35].
This study has some limitations. First, due to the design, the enrolled ICUs and mechanically ventilated patients do not represent all ICUs or patients in any studied country. Second, confounding variables such as severity of illness, presence of comorbidities, duration of mechanical ventilation, and detailed information about intensivists and respiratory therapists were not controlled in this study. Third, missing data for clinical outcomes might have compromised the precision of our multivariable regression analyses.
This study compared the characteristics and ventilation practices of mechanically ventilated patients in Korean ICUs with those in other high-income Asian countries. Korean ICUs were characterized by older patients, higher patient-to-nurse ratios, and more beds per unit, reflecting a greater burden on ICU staff. Despite those challenges, clinical outcomes in Korean ICUs were comparable to those in other high-income Asian countries. Although adherence to many recommended ventilation strategies was observed, mortality predictors highlight the significant effects of age and illness severity. The suboptimal use of sufficient PEEP for ARDS indicates a potential area for improvement. Additionally, the beneficial effects of respiratory therapists on mortality rates in Korean ICUs warrant further investigation.
KEY MESSAGES
▪ Compared to other high-income Asian countries, South Korean patients were significantly older and had a lower mean ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen; however, they had a lower mean Sequential Organ Failure Assessment (SOFA) score.
▪ Despite these differences, the 28-day mortality rate did not differ significantly between South Korea and other high-income Asian countries.
▪ The suboptimal use of sufficient positive end-expiratory pressure for patients for acute respiratory distress syndrome highlights potential areas for improvement, and the beneficial impact of respiratory therapists on mortality warrants further investigation in South Korean intensive care units.
Notes
CONFLICT OF INTEREST
Kwangha Lee and Woo Hyun Cho are editorial board members of the journal but were not involved in the peer reviewer selection, evaluation, or decision process of this article. No other potential conflict of interest relevant to this article was reported.
FUNDING
None.
ACKNOWLEDGMENTS
None.
AUTHOR CONTRIBUTIONS
Conceptualization: YSK. Data curation: KHN, KMJ, SKH, AYL, JHA, HJJ, KSB, SHP, SJP, YMS, JSP, SWK, JHK, JKP, DJK, BYL, WHC, KHL, SIL, TSP, YJJ, SHK, SBJ, SHK, WJJ, SML, SHP, YSS, YJC, YSK. Methodology: KHN, YSK. Writing – original draft: KHN. Writing – review & editing: YSK. All authors read and agreed to the published version of the manuscript.