;on behalf of the Korean Society of Nephrology Disaster Preparedness and Response CommitteeIntroduction
Chronic kidney disease (CKD) is a condition that significantly increases the social burden due to the excessive healthcare costs associated with its advanced stages and the possibility of requiring renal replacement therapy if it progresses to end-stage kidney disease (ESKD) [1]. With the aging of society, the prevalence of chronic diseases such as diabetes and hypertension has recently increased, as has the prevalence of CKD [2]. Although the mortality rates of patients with ESKD have declined in the United States [3] and South Korea [4], CKD has still emerged as a leading cause of mortality worldwide, according to the Global Burden of Disease study [5]. Patients with ESKD visit the emergency department (ED) more frequently than the general population because of complications related to dialysis procedures or underlying conditions. However, there is a lack of research in South Korea that compares the frequency of ED visits, hospitalization rates, and mortality of patients with ESKD with those without CKD and examines the clinical outcomes of patients with ESKD after ED visits. A previous study reported that patients with ESKD had higher rates of ED use and subsequent hospitalization than the national mean rates for adults in the United States [6]. The study identified factors associated with higher rates of ED visits, including younger age, female sex, comorbidities (respiratory abnormalities and congestive heart failure), and preventable factors (hyperkalemia, fluid overload, catheter use, or graft hemodialysis [HD] access).
Previous studies have shown that infection, electrolyte disturbances, and cardiovascular complications are common causes of ED visits in patients with ESKD [6–8]. The unexpectedness of ED visits can have a detrimental impact on outcomes; therefore, understanding the patterns of ED visits and investigating associated factors in this patient population are important for improving clinical outcomes. According to a meta-analysis by Han et al. [8] on the ED visit patterns of HD patients, various factors, such as access to care, the burden of comorbid conditions, and changes in healthcare models, have been identified as predictors of ED use. However, only a limited number of interventions have been effective in reducing ED visits in ESKD patients who are particularly vulnerable during disasters, as disruptions in HD can have life-threatening consequences [9]. In South Korea, the Emergency Medical Act to collect treatment-related information transmitted from emergency medical institutions nationwide in real time, the National Emergency Department Information System (NEDIS), laid the foundation for building an advanced emergency medical system and providing a basis for research and policy formulation for emergency medical care [10,11]. Since 2016, more than 95% of the country’s emergency medical centers have participated in the database, and approximately 400 hospitals have been registered, sending information on patients who visit their EDs to the National Emergency Medical Center [10,11]. NEDIS includes comprehensive information on ED visits across South Korea, such as patient demographics, triage levels, diagnoses, treatments, and outcomes [10].
In this study, we aimed to investigate ED visits in patients with ESKD using a nationwide ED-based patient registry using NEDIS data and to analyze the causes of ED visits and factors associated with clinical outcomes in patients with ESKD.
Methods
Study design and data sources
This retrospective observational study used the NEDIS database to obtain information on ED visits at regional and local emergency medical centers from January 1, 2019 to December 31, 2021. NEDIS is a nationwide database operated by the National Emergency Medical Center in Korea. This study is also part of the Korean Society of Nephrology’s (KSN) cooperative study “Development and Evaluation of a Disaster Response System for Hemodialysis Units” with a 2022-year memorandum of understanding between the National Emergency Medical Center and the KSN Disaster Preparedness and Response Committee. The KSN Disaster Preparedness and Response Committee aimed to prepare measures to respond to traditional disasters such as fires and power outages, as well as strengthen responses to earthquakes, typhoon damage due to climate change, and new infectious diseases [12,13].
This study was conducted in accordance with the principles of the Declaration of Helsinki. The database was fully anonymized, and the requirement for informed consent was waived by the Institutional Review Board of Ulsan University Hospital (No. UUH2024-01-017 and UUH2023-10-034).
Study population
We examined all adult ED visitors aged 20 years and older with ESKD during the study period. ED visits by patients with ESKD were selected if they had a primary or secondary ED diagnosis of N18.5 using the modified version of the International Classification of Diseases, 10th revision (ICD-10). ESKD was defined as stage 5 CKD regardless of whether the patient received HD or peritoneal dialysis. We identified 19,563,278 ED visits, of which patients with CKD stages 1 to 4 were excluded. We also excluded 4,501 visits with no information on the Korean Triage and Acuity Scale (KTAS), ED outcomes, or length of stay. Finally, 125,392 ED visits of patients with ESKD and 19,287,972 ED visits of patients without CKD were included in the study (Fig. 1).
Variables and study outcomes
The data included demographic characteristics (age and sex), insurance status, and information about ED visits. Age was grouped into 10-year increments. The Ministry of Health and Welfare has designed three ED levels, depending on hospital function and size [14]: level 1, regional emergency medical center (REMC); level 2, community emergency medical center (CEMC); and level 3, community emergency medical institute (CEMI). The route of arrival (direct visit, transfer from another hospital, transfer from an outpatient clinic, and others) and length of ED stay were included. The KTAS triage system is used to categorize patients who visit the ED based on their symptoms and applies the patient’s urgency and priority to the ED. The method was developed through the Korean emergency patient severity classification system standardization study from 2012 to 2015 and implemented in all EDs in South Korea in 2016 [15]. The KTAS consists of a five-level system (1, resuscitation; 2, emergency; 3, urgent; 4, less urgent; and 5, nonurgent) that classifies patients based on the severity of symptoms, vital signs, and chief complaints [15,16]. The KTAS score allocated by the ED physicians determines the patient’s waiting time until assessment by an ED doctor [17]. ED outcomes (discharge to home, transfer to another hospital, admission, and death in the ED) were categorized. The causes of ED visits, such as heart disease, cerebrovascular disease, infection, electrolyte imbalance, and vascular access, were identified based on the primary diagnosis of ED visits (Supplementary Fig. 1, available online). The study outcomes included hospitalization after an ED visit, in-hospital mortality, and clinical outcomes after admission.
Statistical analysis
Comparisons between groups were performed using chi-square tests for categorical variables and t tests or analysis of variance for continuous variables. Data are presented as frequencies (percentage) and means (standard deviations). The risks of hospitalization and mortality in ED visits by patients with ESKD were analyzed using logistic regression analyses adjusted for age, sex, insurance status, hospital service level, mode of arrival, KTAS classification, and length of stay. All tests were two-tailed, and a p-value <0.05 was considered statistically significant. Statistical analyses were performed using R version 4.0.5 (R Foundation for Statistical Computing).
Results
Baseline characteristics of the end-stage kidney disease and non-chronic kidney disease groups
Of the 129,893 ED visits of patients with ESKD, 8.0% were in their 40s, whereas 17.8%, 25.7%, 26.3%, and 17.6% were in their 50s, 60s, 70s, 80s, and older, respectively (Table 1). Moreover, 42.4% were male and 23.2% were receiving medical aid, while 39.5% of ED visits were to REMCs and 51.1% were to CEMCs. A total of 24.9% were sent from other hospitals, and 8.5% were sent from outpatient departments. The KTAS classifications at the time of the ED visit were 3.7%, 14.9%, and 56.4%, which indicated resuscitation, emergency, and urgency, respectively. The average length of stay in the ED for the ESKD group was 7.97 hours, with 5.88 and 8.86 hours for discharge and admission, respectively. The non-CKD group showed relatively fewer differences in the frequency of visits according to age group, were more likely to be female, and were less likely to be supported by medical aid. The non-CKD group was more likely to use CEMC or CEMI and present directly to the ED, with a higher proportion of less urgent and nonurgent KTAS classifications. ED visits without CKD had shorter ED stays than those with ESKD.
Emergency department visits in the end-stage kidney disease and non-chronic kidney disease groups
Fig. 2 shows the outcomes of the ED visits. In the ESKD group, 30.8% out of total 125,392 ED visits were discharged, and 66.7% required hospitalization: 49.7% in the general ward, 17.6% in the intensive care unit (ICU), and 0.01% in others. A total of 543 cases (0.4%) died in the ED. Meanwhile, 75.9% out of total 17,153,431 ED visits in the non-CKD group were discharged, while 21.0% were admitted to the hospital: 17.6% in general wards, 3.3% in ICU, and 0.07% in others. Of the ED visits by non-CKD patients, 0.7% died in the ED. There were 2,343 cases in ESKD group (1.9%) and 325,562 cases in non-CKD group (1.9%) that were transferred to other hospitals. Fig. 3 shows the hospitalization rates and in-hospital mortality rates for the ESKD and non-CKD groups. ED visits by ESKD patients had significantly higher hospitalization (66.7%) and in-hospital mortality (9.4%) rates compared to those by non-CKD patients (21.0% and 5.1%, respectively). The in-hospital mortality rate in the ESKD group was substantially higher than that of the non-CKD group after hospitalization. The number of ED visits was 42,213 (33.7%), 40,874 (32.6%), and 42,305 (33.7%) in 2019, 2020, and 2021, respectively, compared to 6,013,361 (35.1%), 5,399,041 (31.5%), and 5,741,029 (33.5%) for the non-CKD group. Hospitalization rates were higher for patients with ESKD in 2021 but higher for those by patients without CKD in 2020. The post-hospitalization mortality rates were higher in 2021 in both groups (Supplementary Fig. 1, available online).
Causes and patterns of emergency department visits among patients with end-stage kidney disease
Among all ED visits for patients with ESKD, vascular access complication (T82) was the most common cause by ICD-10 diagnosis code, accounting for 4.1% of visits, followed by disorder of digestive system at 3.1% (Table 2). ESKD patients are prone to pneumonia (2.9%), pulmonary edema (2.7%) from fluid overload caused by inadequate dialysis, and fluid, electrolyte, and acid-base disorders (2.2%) leading to complications like hyperkalemia or acidosis (Table 2). The in-hospital mortality rate following hospitalization was notably high for ED visits due to pneumonia, whereas it remained low for visits related to vascular access problems, highlighting significant variability in outcomes depending on the cause of the ED visit.
Clinical outcomes of emergency department visits among patients with end-stage kidney disease
Table 3 presents the clinical outcomes of post-hospitalization ED visits for patients with ESKD. Overall, 74.6% were discharged, 13.7% were transferred, and the mortality rate was 9.4%. Of the cases admitted, 61,884 (74.4%) were admitted to the general ward and 21,242 (25.6%) to the ICU. Of the ED visits admitted to general wards, 80.2% were discharged, 11.8% were transferred to other hospitals, and 5.9% died. In contrast, 58.4% of the ICU admissions were discharged, 19.2% were transferred, and 19.5% died.
Factors associated with hospitalization and in-hospital mortality in patients with end-stage kidney disease
Multivariable regression analysis of the likelihood of hospitalization after an ED visit for patients with ESKD showed that older age, male sex, transfer from another hospital or outpatient clinic, higher level of hospital service, KTAS classification severity, and longer ED stay were significantly associated (Table 4). Hospitalization rates were lower for cases with medical aid than for those covered by the National Health Insurance and higher for those with other insurance. Table 5 presents the factors associated with increased mortality, with results similar to those for hospitalization. Older age, male sex, transfer from another hospital, severe KTAS classification, and longer hospital stay were associated with higher post-hospital mortality among patients with ESKD. In addition, the mortality rate was higher when patients were admitted to CEMC or CEMI than when they were admitted to REMC.
Discussion
This study is the first to use nationwide ED visit data to examine the utilization patterns of patients with ESKD in South Korea. The study revealed that ED visits by Korean patients with ESKD included a higher proportion of older adults and female and a higher proportion of patients who received medical aid. These patients tended to arrive at the ED with high KTAS classification severity and were more likely to visit high-level centers. Patients with ESKD who visited the ED were more likely to be transferred from other hospitals or outpatient settings and to have longer ED stays than those without CKD. Compared to patients with ESKD who visited the ED, those without CKD were more evenly distributed in age and had a higher proportion of patients with low-severity conditions on arrival at the ED. However, a study of ED visit trends using NEDIS data in the Korean general population found that as KTAS scores decreased (i.e., as severity increased), the proportion of patients who used a 119 ambulance, were admitted to the ICU, or died in the ED increased [10]. Patients with ESKD had a higher proportion of patients with multiple comorbidities and lower KTAS scores upon ED arrival, which may explain their poorer prognosis compared to the non-CKD group.
A recent study identifying the epidemiological characteristics and trends of older patients presenting to EDs in Korea found that elderly patients had a higher proportion of female patients, more hospitalizations, higher severity of presenting conditions, and higher ED and in-hospital mortality rates [18]. The authors reported that 36.1% of elderly patients were hospitalized, and 18.3% of the hospitalized patients were admitted to the ICU. The ED and in-hospital mortality rates were 1.8% and 4.6%, respectively. In this study, the patients with ESKD who visited the ED also had a higher proportion of older patients and were similar to those in a previous study regarding patient characteristics and clinical course. However, this study found higher rates of hospitalization and in-hospital mortality, which is expected, as patients with ESKD tend to have many comorbidities. Patients with ESKD tended to use higher levels of hospital services, with 25.5% of hospitalized patients admitted to the ICU. The current study focused on investigating the utilization patterns of EDs by individuals with ESKD, and the clinical outcomes of national ED visits in Korea. These findings can serve as a basis for developing and improving healthcare delivery policies for patients with CKD, a growing population with an increasing number of chronic diseases.
ESKD is a complex clinical condition requiring special care in both emergency and nonemergency cases. Patients with ESKD use the ED more frequently than the general population. A recently published population-based study of a U.S. Medicare cohort found that patients with ESKD visited the ED six times more often than the general population [6]. Yoo et al. [10] investigated the trends in ED visits among the Korean general population using NEDIS data from 2018 to 2022 and found that the standardized number of ED visits per 100,000 people was approximately 18,000 to 20,000 during the study period. The frequency of ED visits in patients with ESKD compared to the general population in this study appears to be lower than that reported in other countries; however, this may be due to differences in healthcare conditions and regional characteristics among countries. Understanding the burden of conditions is crucial to evaluating ED use among ESKD patients. This study revealed that common reasons for ED visits in the ESKD group included complications of vascular devices, digestive system disorders, pneumonia, pulmonary edema, and electrolyte or acid-base imbalances. Hospitalization rates were higher for conditions like pneumonia and pulmonary edema, which are also associated with increased mortality after admission (Table 2).
Although this study included the coronavirus disease 2019 (COVID-19) pandemic period, it did not provide a comprehensive analysis of COVID-19-related ED visits for patients with ESKD in South Korea. A previous study by Na et al. [19] examined the impact of the COVID-19 pandemic on hospitalization and excess ED mortality in patients with acute myocardial infarction, stroke, and severe trauma in South Korea, revealing fewer ED visits and more ED deaths in patients with these three leading emergency conditions. Moreover, the COVID-19 pandemic led to social distancing measures, which likely influenced the healthcare-seeking behaviors of patients with ESKD. This factor should be considered when interpreting our findings on the frequency and severity of ED visits among patients with ESKD. This study found that ED visits in 2020 tended to decrease in both the ESKD and non-CKD groups compared to those in 2019. Hospitalization and mortality rates tended to increase in patients with ESKD after the COVID-19 outbreak in 2020, whereas only mortality rates tended to increase in the non-CKD group (Supplementary Fig. 1, available online). These modest changes in hospitalization and mortality rates are consistent with findings from studies of the epidemiologic trends of ED visits in South Korea [10].
This study identified several factors associated with higher hospitalization (Table 4) and in-hospital mortality (Table 5) rates among patients with ESKD. Older age, male sex, transfer from another hospital or outpatient clinic, higher severity on the KTAS, and longer ED stay were significantly associated with these adverse outcomes, highlighting the complex health profiles and critical conditions of patients with ESKD who present to the ED [8]. Understanding these associations is crucial for developing targeted interventions to improve patient outcomes. For instance, early identification and management of patients with high-risk ESKD in outpatient settings could reduce the need for emergency care and subsequent hospitalizations [8]. The KTAS, which is used to categorize patients according to the urgency of their condition, plays a pivotal role in assessing the severity of patients with ESKD upon arrival at the ED, which is consistent with previous studies [15]. The high proportion of patients with ESKD classified as having severe KTAS levels underscores the critical nature of their health issues. Implementing more robust triage protocols and ensuring quick access to necessary treatments could help manage these patients more effectively and reduce their high hospitalization and mortality rates.
This study has several limitations to disclose. The retrospective study design may have introduced biases related to data recording and patient selection. Additionally, the study relied on administrative data, which may have lacked detailed clinical information that could influence outcomes for dialysis patients and could not obtain procedure codes to distinguish between peritoneal dialysis and HD, so our findings are constrained by the use of the N18.5 code to define patients with ESKD. The inability to further subclassify N18.5 limits our ability to analyze patients with non-dialysis ESKD or those with failed transplants in greater detail. This limitation should be considered when interpreting the results of our study. We separated the ESKD group from the non-CKD group based on the ED discharge diagnoses provided. It is possible that N185 was included in the discharge diagnosis for ESKD to allow for the special exemption code, but it is also possible that the discharge diagnosis for patients with CKD stages 1 to 4 did not include a CKD diagnosis. This group is based on a maximum of 20 discharge diagnoses, but the non-CKD group may include patients with CKD stages 1 to 4. Furthermore, the impact of COVID-19 and related healthcare disruptions on ED visit patterns was not explicitly analyzed.
In conclusion, this study highlights the critical healthcare needs of patients with ESKD in South Korea who frequently visit the ED due to severe conditions, leading to high hospitalization and mortality rates. Addressing the identified risk factors through targeted interventions, improving triage protocols, and considering the broader impacts of public health measures, such as social distancing, could enhance care for this vulnerable population. These findings underscore the need for tailored healthcare policies to improve the clinical outcomes of patients with ESKD, ultimately aiming to reduce the burden on emergency services and improve patients’ quality of life.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement figure 1
Print
