Epidemiology and Functional Outcome of Acute Stroke Patients in Korea Using Nationwide data

Article information

J Korean Neurosurg Soc. 2025;68(2):159-176

Publication date (electronic) : 2024 September 24

doi : https://doi.org/10.3340/jkns.2024.0118

Seungmin Shin¹, Young Woo Kim², Seung Hun Sheen³, Sukh Que Park⁴, Sung-Chul Jin⁵, Jin Pyeong Jeon⁶, Ji Young Lee², Boung Chul Lee⁷, Young Wha Lim⁷, Gui Ok Kim⁷, Youg Uk Kwon⁸, Yu Ra Lee⁹, So Young Han¹⁰, Jae Sang Oh^,2

¹Department of Neurosurgery, Soonchunhyang University Cheonan Hospital, College of Medicine, Soonchunhyang University, Cheonan, Korea

²Department of Neurosurgery, Uijeongbu St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea

³Department of Neurosurgery, Bundang CHA Medical Center, CHA University, Seongnam, Korea

⁴Department of Neurosurgery, Soonchunhyang University Seoul Hospital, College of Medicine, Soonchunhyang University, Seoul, Korea

⁵Department of Neurosurgery, Inje University Haeundae Paik Hospital, Busan, Korea

⁶Department of Neurosurgery, Hallym University College of Medicine, Chuncheon, Korea

⁷Health Insurance Review & Assessment Service (HIRA), Wonju, Korea

⁸Healthcare Review and Assessment Committee (HCRAC), Seoul, Korea

⁹Cardio-Cerebrovascular Disease Assessment Division, Quality Assessment Administration Department, Healthcare Review and Assessment Committee (HCRAC), Seoul, Korea

¹⁰Quality Assessment Management Division, Quality Assessment Department , Healthcare Review and Assessment Committee (HCRAC), Seoul, Korea

Address for reprints : Jae Sang Oh Department of Neurosurgery, Uijeonbu St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 271 Cheonbo-ro, Uijeongbu 11765, Korea Tel : +82-31-820-3623, Fax : +82-31-820-3618, E-mail : metatron1324@hotmail.com

Received 2024 July 14; Revised 2024 August 30; Accepted 2024 September 21.

Abstract

Graphical Abstract

INTRODUCTION

Stroke remains a significant health issue, with a high mortality up to 30% within the first year. Survivors often face severe disabilities and long-term complications, leading to increased long-term mortality. With the rapid aging of the population, stroke imposes substantial social and economic burdens on South Korea. Continuous national efforts have led to a gradual decrease in the age-standardized stroke mortality rate, from 76.5 per 100000 in 2012, following the implementation of Acute Stroke Acessment Registry since 2006, and it was decreased to 3th most common cause of death in Korea [4]. Despite these efforts, stroke remains the fifth leading cause of death, following cancer and heart disease, comparable to the OECD member countries’ average standardized mortality rates. Therefore, ongoing national efforts and medical research are essential to further reduce stroke mortality.

Due to variations in race and nationality, there is a need for epidemiological studies on the proportions of subtypes of strokes such as subarachnoid hemorrhage (SAH), intracerebral hemorrhage (ICH), and acute ischemic stroke. Generally, ischemic stroke is known to be over twice as common as hemorrhagic stroke. However, research indicates that asians tend to have relatively smaller vessel diameters, a higher prevalence of small vessel disease, and increased rates of intracranial arterial stenosis. Therefore, there is a necessity for epidemiological research on the subtype proportions of stroke in Korea.

To improve the quality of stroke care and reduce the mortality, the Acute Stroke Acessment Registry was initiated in 2006 by government. It includes data from hospitals with at least 10 admissions of acute stroke patients per year, involving 220 multi-centers nationwide. Using this registry, we were able to confirm recent changes in mortality rates and treatment methods for ischemic stroke and ICH [8]. We were able to observe differences in treatment outcomes for ICH or SAH based on hospital size in Korea, as well as changes in treatment trends for acute ischemic stroke [4,6,7]. Previous studies utilizing this registry have been conducted in the majority of stroke treatment hospitals nationwide, and the patient data is highly accurate. Therefore, conducting recent epidemiological studies on stroke using this registry is highly justified.

Korea boasts excellent medical accessibility and operates a nationwide health insurance system. As part of the four major serious illness management project, acute stroke has been included since 2013, expanding insurance benefits to encourage aggressive treatment. Therefore, upon admission to the emergency room for most strokes, thorough examinations and treatments are recommended. This swift early treatment improves patient prognosis. The Acute Stroke Assessment Registry records patient outcomes upon discharge. Through this, we can understand the prognosis of patients with acute stroke. Using data from the Acute Stroke Assessment Registry, we investigated the epidemiology and functional outcomes of stroke in Korea.

MATERIALS AND METHODS

The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Research Ethics Committee of The Catholic University of Korea Uijeongbu St. Mary’s Hospital (IRB number. UC23ZISE0023). This study was performed under the joint project of the National Stroke Quality Assessment Research with the National Health Insurance Administration. The requirement for informed consent was waived due to the retrospective nature of the study.

This study merged the Acute Stroke Acessment Registry and the National Health Insurance Service’s data to analyze the current status in Korea. The Acute Stroke Acessment Registry was designed to facilitate ongoing quality management of stroke medical services, aiming to reduce mortality and disability rates and improve the quality of medical services in care facilities. Since the initial approval of the acute stroke primary evaluation by the Ministry of Health and Welfare in 2006, data have been collected continuously through periodic evaluations.

In order to manage the quality of stroke medical services, a total of 24 indicators, including nine evaluation indicators and 15 monitoring indicators, are collected prospectively every 2 years for a period of 3–6 months. The target hospitals include tertiary general hospitals and general hospitals with more than 10 admissions for acute stroke, with approximately 220 hospitals nationwide participating. Over the past year, patients admitted to the emergency room within 7 days of symptom onset with a primary diagnosis of I60-63 without the specified comorbidities were included in the study.

We conducted an analysis of emergency room visits by time after symptom onset using merged data. Subsequently, we analyzed the treatment status of patients after hospital arrival and examined treatment outcomes based on whether the patients were admitted to tertiary hospitals or general hospitals.

The National Health Insurance Service evaluated the quality of acute stroke care quarterly from 2007 to 2014 and semi-annually from 2016 onwards. Evaluation targeted acute stroke patients admitted through the emergency room within 7 days of symptom onset, utilizing de-identified data to ensure patient privacy. For this study, we analyzed adequacy evaluation data during 18 months from 2016, 2018, and between 2020 to 2021. The study population comprised 86568 patients admitted for acute stroke to tertiary and general hospitals nationwide. Key diagnoses included ICH (I60), SAH (I61), and ischemic stroke (I63), with only patients with clear symptom onset times included.

The data consisted of patient surveys and institutional surveys. The patient survey contained demographic information, diagnosis, admission/discharge details, and medical history, while the institutional survey included hospital status, specialist counts, and stroke unit availability. We merged and analyzed these datasets. Stroke severity was assessed using the National institutes of health stroke scale (NIHSS) and Glasgow coma scale (GCS). NIHSS scores range from 0 (normal) to 42 (most severe), while GCS scores range from 15 (clear consciousness) to 3 (most severe). Both NIHSS and GCS scores were categorized for analysis.

Patient discharge prognosis measurement involves analyzing records of patients who are assigned scores ranging from 0 to 100 on the Korean version of modified barthel index (K-MBI), modified Barthel index (MBI), and Barthel index (BI), and from 18 to 126 on the Functional independence measure (FIM). Additionally, the modified Rankin scale (mRS), which classifies patients into six grades ranging from normal (grade 0) to deceased (grade 6), and the Glasgow outcome scale (GOS), which categorizes patients from grade 5 (normal) to grade 1 (poor prognosis), are recorded. Patient discharge prognosis is determined based on predefined criteria for each scale. Patients scoring 75 or higher on K-MBI, MBI, or BI, and 90 or higher on FIM, are categorized as having a good prognosis. For mRS, patients scoring 2 or below are considered to have a good prognosis, while for GOS, a grade of five indicates a good prognosis.

Demographic characteristics and disease status were analyzed for each year. Statistical analysis was performed using SAS (Statistical Analysis System) software version 9.4 (SAS Institute, Cary, NC, USA), and the distribution of patient groups by year was presented. Basic characteristics of patient groups were reported as population (mean±standard deviation) or count (percentage).

RESULTS

Characteristics of acute stroke patients

Table 1 summarizes the demographic distribution of 86568 acute stroke patients included for 18 months from 2016 to 2021. Among a total of 86568 stroke patients, 7218 (8.3%) had a diagnosis of SAH, 14119 (16.3%) had ICH, 1846 (2.1%) had non-traumatic subdural hemorrhage, and 64841 (74.9%) had ischemic stroke.

Table 1.

Demographic distribution of acute stroke patients during 18 months from 2016 to 2021

Analysis of patient distribution by stroke classification revealed that SAH patients numbered from July 2016 to December 2016 : 2348 (8.8%), from July 2018 to December 2018 : 2417 (8.3%), from October 2020 to March 2021 : 2453 (7.9%), with a decreasing trend each year. Conversely, for ICH patients, the numbers were 4246 (16.0%) from July 2016 to December 2016, 4664 (16.0%) from July 2018 to December 2018, 5209 (16.9%) from October 2020 to March 2021, showing an increasing trend. There was little variation among ischemic stroke patients, with numbers as follows : from July 2016 to December 2016 : 19816 (74.5%), from July 2018 to December 2018 : 21930 (75.4%), from October 2020 to March 2021 : 23095 (74.7%).

Analysis of patient distribution by stroke severity assessment tool among all acute stroke patients showed that 66247 (76.5%) were evaluated using NIHSS, 18933 (21.9%) using GCS, and 1388 (1.6%) using other assessment tools.

In this study, patients with clear symptom onset time evaluated for stroke severity using NIHSS or GCS were targeted. Among patients with clear symptom onset time, totaling 59152 (68.3%), the distribution for each grade was as follows : from July 2016 to December 2016 : 18530 (69.7%), from July 2018 to December 2018 : 20289 (69.8%), from October 2020 to March 2021 : 20333 (65.8%).

Table 2 presents the baseline characteristics of acute stroke patients. Among a total of 59152 patients with clear symptom onet time, 33595 (56.8%) were male and 25557 (43.2%) were female. Patients aged between 18 and 45 years accounted for 4041 (6.8%), those aged between 46 and 59 years accounted for 13833 (23.4%), those aged between 60 and 69 years accounted for 13604 (23.0%), and those aged 70 years or older accounted for 27674 (46.8%).

Table 2.

Baseline characteristics of acute stroke patients with clear onset time

Among patients undergoing severity assessment, 44463 (75.2%) underwent NIHSS evaluation, while 14689 (24.8%) underwent GCS assessment. Analysis categorized by NIHSS scores revealed that patients with scores between 0 and 4 were 28022 (63.0%), between 5 and 7 were 7252 (16.3%), between 8 and 13 were 4818 (10.8%), between 14 and 21 were 3021 (6.8%), between 22 and 42 were 1350 (3.0%). Thirty-four thousand four hundred twenty-six (58.2%) utilized emergency medical services.

In this study, we analyzed the distribution of patient arrival times to the hospital by time intervals following the onset of symptoms. Table 3 presents the distribution of patients by time intervals after symptom onset. Among all patients, 12218 (20.7%) arrived at the hospital within 1 hour of symptom onset, while 7902 (13.4%) arrived between 1 and 2 hours after symptom onset. Additionally, 4800 patients (8.1%) arrived between 2 and 3 hours after symptom onset, and 3231 patients (5.5%) arrived between 3 and 4 hours after symptom onset. Moreover, 2525 patients (4.3%) arrived between 4 and 5 hours after symptom onset. Furthermore, among patients arriving at the hospital within 4.5 hours of symptom onset, there were 9441 patients (50.9%) from July 2016 to December 2016, 10093 patients (49.7%) from July 2018 to December 2018, and 9800 patients (48.2%) from October 2020 to March 2021. Overall, 29334 patients (49.6%) in the entire acute stroke population in Korea arrived at the hospital within 4.5 hours of symptom onset.

Table 3.

Distribution, hospital characteristics, discharge outcomes of acute stroke patients

Table 3 combines data from the acute stroke patient survey and institutional survey to investigate the characteristics of hospitals where patients received treatment over the years. Of the total patients, 27332 (46.2%) received treatment at tertiary hospitals, while 31820 (53.8%) received treatment at general hospitals. Patients treated at tertiary hospitals were 8695 (46.9%) from July 2016 to December 2016, 10026 (49.4%) from July 2018 to December 2018, and 8611 (42.3%) from October 2020 to March 2021, showing a decreasing trend each year. Conversely, patients treated at general hospitals were 9835 (53.1%) from July 2016 to December 2016, 10263 (50.6%) from July 2018 to December 2018, and 11722 (57.7%) from October 2020 to March 2021, showing an increasing trend each year.

Among the patients, 38999 (65.9%) received treatment at hospitals with stroke unit facilities, while 20153 (34.1%) received treatment at hospitals without stroke unit facilities.

According to the survey results on hospital manpower, the number of neurologists was 6.2±5.10 from July 2016 to December 2016, 6.61±5.41 from July 2018 to December 2018, and 6.24±5.33 from October 2020 to March 2021. The number of neurosurgeons was 7.17±4.98 from July 2016 to December 2016, 7.01±5.00 from July 2018 to December 2018, and 6.68±4.76 from October 2020 to March 2021. The number of rehabilitation medicine specialists was 3.28±2.28 from July 2016 to December 2016, 3.31±2.46 from July 2018 to December 2018, and 3.12±2.41 from October 2020 to March 2021.

The discharge outcomes of patients with acute stroke were analyzed annually. Table 3 summarizes the annual discharge outcomes of patients with acute stroke. Among the total acute stroke patients, discharge outcomes were recorded for 57998 patients (98.0%). Among them, 33784 patients (58.3%) were discharged with good outcomes, while 24214 patients (41.7%) were discharged with poor outcomes.

Charateristics of SAH patient

Among a total of 7218 patients with SAH, 1439 patients (19.9%) were evaluated for severity using NIHSS, while 5583 patients (77.3%) were evaluated using GCS, and 196 patients (2.7%) were evaluated using other assessment criteria.

Among patients evaluated for severity using NIHSS or GCS, a total of 5850 patients (81.0%) with clear symptom onset time were identified as having SAH, including 1934 patients (82.4%) from July 2016 to December 2016, 1967 patients (81.4%) from July 2018 to December 2018, and 1949 patients (79.5%) from October 2020 to March 2021.

Table 4 presents the analysis results of the baseline characteristics of SAH patients with clear symptom onset time. Among a total of 5850 patients, 2134 (36.5%) were male, and 3716 (63.5%) were female. Patients aged between 18 and 45 years accounted for 956 (16.3%), those between 46 and 59 years accounted for 2281 (39.0%), those between 60 and 69 years accounted for 1242 (21.2%), and those aged 70 years or older accounted for 1371 (23.4%).

Table 4.

Baseline characteristics, hospital characteristics, discharge outcomes of subarachnoid hemorrhage patients

During the severity assessment, 1169 (20.0%) SAH patients underwent NIHSS evaluation, while 4681 (80.0%) underwent GCS evaluation. When categorized based on NIHSS scores, patients with scores from 0 to 4 accounted for 670 (11.5%), those with scores from 5 to 7 accounted for 67 (1.1%), those with scores from 8 to less than 13 accounted for 100 (1.7%), those with scores from 14 to less than 21 accounted for 99 (1.7%), and those with scores from 22 to less than 42 accounted for 233 (4.0%). When categorized based on GCS scores, patients with scores from 13 to less than 15 accounted for 3055 (52.2%), those with scores from 9 to less than 12 accounted for 416 (7.1%), and those with scores from 0 to less than 8 accounted for 1210 (20.7%).

Among all patients, 4693 (80.2%) arrived at the hospital via ambulance, while 1157 (19.8%) used other means of transportation.

The characteristics of hospitals where patients with SAH received treatment were investigated (Table 4). Among them, 3034 patients (51.9%) were treated at tertiary comprehensive hospitals, while 2816 patients (48.1%) received treatment at general hospitals. The number of patients treated at tertiary comprehensive hospitals decreased annually, 1083 (56.0%) from July 2016 to December 2016, 1052 (53.5%) from July 2018 to December 2018, 899 (46.1%) from October 2020 to March 2021. Conversely, the proportion of patients treated at general hospitals increased annually, 851 (44.0%) from July 2016 to December 2016, 915 (46.5%) from July 2018 to December, 1050 (53.9%) from October 2020 to March 2021.

Among the patients, 4106 (70.2%) received treatment at hospitals with stroke unit facilities, while 1744 (29.8%) were treated at hospitals without such facilities.

According to the survey results on hospital staff, the number of neurologists was 6.40±4.22 from July 2016 to December 2016, 6.59±4.50 from July 2018 to December 2018, and 6.29±4.70 from October 2020 to March 2021. The number of neurosurgeons was 7.48±4.42 from July 2016 to December 2016, 7.26±4.27 from July 2018 to December 2018, and 7.04±4.61 from October 2020 to March 2021. The number of rehabilitation medicine specialists was 3.38±2.00 from July 2016 to December 2016, 3.42±2.18 from July 2018 to December 2018, and 3.33±2.35 from October 2020 to March 2021.

The discharge outcomes of patients with SAH were analyzed annually. Table 4 summarizes the annual discharge outcomes of SAH patients. Out of the total 5704 SAH patients, discharge outcomes were recorded for 5704 patients (97.5%). Among them, 3191 patients (55.9%) had good outcomes at discharge, while 2513 patients (44.1%) had poor outcomes at discharge.

Among the SAH patients with a clear symptom onset time, 3153 patients (53.9%) had their discharge outcomes measured using the mRS. The mRS scores were as follows : 667 patients (21.2%) scored 0, 1187 patients (37.6%) scored 1, 397 patients (12.6%) scored 2, 225 patients (7.1%) scored 3, 268 patients (8.5%) scored 4, 257 patients (8.2%) scored 5, and 152 patients (4.8%) scored 6.

For those whose outcomes were measured using the GOS, 1125 patients (19.2%) had their outcomes recorded. The GOS scores were as follows : 94 patients (8.4%) scored 1, 62 patients (5.5%) scored 2, 105 patients (9.3%) scored 3, 172 patients (15.3%) scored 4, and 692 patients (61.5%) scored 5.

Charateristics of ICH

Among a total of 14119 patients with ICH, 3424 patients (24.3%) were evaluated for severity using NIHSS, while 10161 patients (72.0%) were evaluated using GCS, and 534 patients (3.8%) were evaluated using other assessment criteria.

Among patients evaluated for severity using NIHSS or GCS, a total of 10403 patients (73.7%) with clear symptom onset time were identified as having ICH, including 3255 patients (76.7%) from July 2016 to December 2016, 3518 patients (75.4%) from July 2018 to December 2018, and 3630 patients (69.7%) from October 2020 to March 2021.

Table 5 presents the baseline characteristics of ICH patients among the total stroke population. Among 10403 patients with clear symptom manifestation, 5768 (55.4%) were male and 4635 (44.6%) were female. Patients aged between 18 and 45 years accounted for 1121 (10.8%), those aged between 46 and 59 years were 3146 (30.2%), patients aged between 60 and 69 years were 2273 (21.8%), and those aged 70 years and older were 3868 (37.1%).

Table 5.

Baseline characteristics, hospital characteristics, discharge outcomes of intracerebral hemorrhage patients

Regarding severity assessment, 2579 (24.8%) ICH patients underwent NIHSS evaluation, while 7824 (75.2%) underwent GCS evaluation. Categorizing NIHSS scores, patients with scores between 0 and 4 were 910 (8.7%), 5 to 7 were 313 (3.0%), 8 to less than 13 were 537 (5.2%), 14 to less than 21 were 411 (4.0%), and 22 to less than 42 were 408 (3.9%). Based on GCS scores, patients with scores between 13 and 15 were 4601 (44.2%), 9 to less than 12 were 1301 (12.5%), and 0 to less than 8 were 1922 (18.5%).

Among all patients, 8108 (77.9%) arrived at the hospital via ambulance, while 2295 (22.1%) used other means of transportation.

The characteristics of hospitals where patients with ICH received treatment were investigated (Table 5). Out of the total, 4918 patients (47.3%) were treated at tertiary hospitals, while 5485 patients (52.7%) were treated at general hospitals. Among those treated at tertiary hospitals, 1546 patients (47.5%) were in July 2016 to December 2016, 1811 patients (51.5%) were in July 2018 to December 2018, and 1561 patients (43.0%) were in October 2020 to March 2021. On the other hand, among those treated at general hospitals, 1709 patients (52.5%) were in July 2016 to December 2016, 1707 patients (48.5%) were in July 2018 to December 2018, and 2,069 patients (57.0%) were in October 2020 to March 2021.

Among the patients, 6959 (66.9%) received treatment at hospitals with stroke care units, while 3444 (33.1%) received treatment at hospitals without stroke care units.

According to the survey results on hospital staffing, the number of neurologists was 5.69±4.03 from July 2016 to December 2016, 6.12±4.35 in the from July 2018 to December 2018, and 5.94±4.32 in the from October 2020 to March 2021. The number of neurosurgeons was 6.83±4.26 from July 2016 to December 2016, 6.64±4.03 from July 2018 to December 2018, and 6.55±4.25 from October 2020 to March 2021. The number of rehabilitation medicine specialists was 3.05±1.96 from July 2016 to December 2016, 3.11±2.07 from July 2018 to December 2018, and 3.04±2.22 from October 2020 to March 2021.

The discharge outcomes of ICH patients were analyzed on a yearly basis. Table 5 summarizes the yearly discharge outcomes of ICH patients. Out of the total ICH patients, 10172 (97.8%) had their discharge outcomes recorded. Among them, 3598 patients (34.6%) were discharged with a favorable outcome, while 6574 patients (64.6%) were discharged with an unfavorable outcome.

Among ICH patients with clear symptom onset times, 5447 patients (52.4%) had their discharge outcomes measured using the mRS. The distribution of mRS scores was as follows : 423 patients (7.8%) scored 0, 1259 patients (23.1%) scored 1, 839 patients (15.4%) scored 2, 716 patients (13.1%) scored 3, 1206 patients (22.1%) scored 4, 808 patients (14.8%) scored 5, and 196 patients (3.6%) scored 6.

Additionally, 1855 patients (17.8%) had their discharge outcomes measured using the GOS. The distribution of GOS scores was as follows : 101 patients (5.4%) scored 1, 137 patients (7.4%) scored 2, 440 patients (23.7%) scored 3, 493 patients (26.6%) scored 4, and 684 patients (36.9%) scored 5.

Charateristics of ischemic stroke patients

Among a total of 64841 ischemic stroke patients, 61548 (94.9%) were evaluated for severity using NIHSS, while 2691 (4.2%) were evaluated using GCS, and 602 (0.9%) were evaluated using other assessment criteria.

Among patients evaluated for severity using NIHSS or GCS, those with a clear onset time of symptoms accounted for a total of 42684 (69.4%), with 13221 (70.5%) from July 2016 to December 2016, 14788 (70.4%) from July 2018 to December 2018, and 14675 (67.4%) from October 2020 to March 2021.

Table 6 presents the baseline characteristics of ischemic stroke patients among the overall stroke population. Among a total of 42684 patients with clearly manifested symptoms of ICH, 25462 (59.7%) were male and 17222 (40.3%) were female. Patients aged between 18 and 45 accounted for 1999 (4.7%) individuals, those between 46 and 59 years old accounted for 8426 (19.7%), those between 60 and 69 years old accounted for 10029 (23.5%), and those aged 70 and older accounted for 22230 (52.1%) individuals.

Table 6.

Baseline characteristics, hospital characteristics, discharge outcomes of ischemic stroke patients

Among ischemic stroke patients undergoing severity assessment, 40809 (95.6%) underwent NIHSS evaluation, while 1875 (4.4%) underwent GCS evaluation. Categorized analysis of NIHSS scores revealed that 26429 (61.9%) patients scored between 0 and 4 points, 6899 (16.2%) scored between 5 and 7 points, 4216 (9.9%) scored between 8 and 13 points, 2528 (5.9%) scored between 14 and 21 points, and 737 (1.7%) scored between 22 and 42 points. Categorized analysis of GCS scores showed that 1506 (3.5%) patients scored between 13 and 15 points, 225 (0.5%) scored between 9 and 12 points, and 144 (0.3%) scored between 0 and 8 points.

Among all patients, 21657 (50.7%) arrived at the hospital via ambulance, while 21027 (49.3%) used other means of transportation.

The characteristics of the hospitals where ischemic stroke patients were treated were investigated annually (Table 6). A total of 19395 patients (45.4%) were treated at tertiary hospitals, while 23289 patients (54.6%) were treated at general hospitals. Among those treated at tertiary hospitals, there were 6065 patients (45.9%) from July 2016 to December 2016, 7195 patients (48.7%) from July 2018 to December 2018, and 6135 patients (41.8%) from October 2020 to March 2021. Among those treated at general hospitals, there were 7156 patients (54.1%) from July 2016 to December 2016, 7593 patients (51.3%) from July 2018 to December 2018, and 8540 patients (58.2%) from October 2020 to March 2021.

A total of 27865 patients (65.3%) were treated in hospitals with stroke units, while 14819 patients (34.7%) were treated in hospitals without stroke units.

Regarding hospital staffing, the number of neurology specialists was 6.29±5.41 from July 2016 to December 2016, 6.72±5.69 from July 2018 to December 2018, and 6.28±5.60 from October 2020 to March 2021. The number of neurosurgery specialists was 7.21±5.20 from July 2016 to December 2016, 7.06±5.25 from July 2018 to December 2018, and 6.66±4.89 from October 2020 to March 2021. The number of rehabilitation medicine specialists was 3.33±2.38 from July 2016 to December 2016, 3.34±2.55 from July 2018 to December 2018, and 3.10±2.46 from October 2020 to March 2021.

The prognosis at discharge for patients with ischemic stroke was analyzed annually. Table 6 summarizes the discharge outcomes for ischemic stroke patients by year. Out of the total ischemic stroke patients, the discharge prognosis was recorded for 41999 patients (98.4%). Among them, 26792 patients (63.8%) were discharged with a good prognosis, while 15207 patients (36.2%) were discharged with a poor prognosis.

Among the ischemic stroke patients with a clear symptom onset time, the discharge prognosis was measured using the mRS for 32667 patients (76.5%). The mRS scores at discharge were as follows : 0 points for 5524 patients (16.9%), 1 point for 11234 patients (34.4%), 2 points for 6136 patients (18.8%), 3 points for 4253 patients (13.0%), 4 points for 3553 patients (10.9%), 5 points for 1787 patients (5.5%), and 6 points for 180 patients (0.6%).

DISCUSSION

In this study, using data from the assessment of Acute Stroke Acessment Registry, we examined the characteristics of domestic acute stroke patients and the characteristics of stroke subtypes. The distribution of specific types of stroke patients in Korea was similar to the global stroke statistics from the American Heart Association/American Stroke Association (AHA/ASA) in 2021, with SAH, ICH, and ischemic stroke accounting for 8.3%, 20.4%, and 76.1% of the total, respectively, which were close to 8.3%, 16.3%, and 74.9%, respectively. The incidence rate of acute stroke patients in Korea showed a similar trend to the global distribution of stroke patients reported by AHA/ASA [9]. The incidence rate of acute stroke patients in Korea is similar to the global distribution of stroke patients reported by AHA/ASA, but the rate of ICH is slightly higher. According to previous studies, the proportion of hemorrhagic strokes in Asia has been shown to be decreasing, but relatively higher. This is presumed to be associated with the control of hypertension [14,15].

Examining the characteristics of stroke patients in Korea, it was found that overall stroke distribution had a higher proportion of males compared to females (33595 [56.8%] vs. 25557 [43.2%]), with a similar pattern observed for ICH (5768 [55.4%] vs. 4635 [44.6%]) and ischemic stroke (25462 [59.7%] vs. 17222 [40.3%]). However, for SAH, the proportion of females was higher than males (2134 [36.5%] vs. 3716 [63.5%]). Female gender has been recognized as a risk factor for SAH and has been mentioned as a risk factor in several studies [2]. In a previous study analyzing 83 study populations, women had a higher incidence of cerebral aneurysms than men, with a higher number of aneurysms larger than 5 mm in the female group. Since larger aneurysms pose a greater risk of rupture, the higher incidence of SAH in women observed in previous studies reflects a similar improvement [2,10,13]. Secondly, in animal experiments on the formation of cerebral aneurysms in estrogen and estrogen receptor in two different areas of the brain, the incidence of aneurysms in rats undergoing ovariectomy was significantly higher, and menopause in women was shown to increase the incidence of aneurysms [13]. Additionally, conditions such as preeclampsia, eclampsia, hemolysis, elevated liver enzymes, and low platelet count syndrome accompanying pregnancy-induced hypertension are the most important stroke risk factors in women, particularly the most significant stroke risk factor for pregnant women [10]. The stroke risk factor for pregnant women seems to have been reflected in the results, and women generally have a higher incidence of SAH four to 5 years after menopause, which is thought to increase the incidence of SAH in women due to the relative loss of estrogen and progesterone regulating inflammation and oxidative stress following menopause and the relative loss of estrogen preventing the formation of cerebral aneurysms [12]. In addition to higher prevalence of hypertension, genetic factors are also known to contribute to the higher incidence of SAH in women [2,10,13].

Furthermore, the distribution of patient ages varied by stroke subtype. For SAH, patients under 59 years old accounted for more than half of the total patients (3237, 55.3%). As previously revealed in existing research, SAH presumed to be influenced by hormonal changes following postmenopausal, such as estrogen, occurs before the age of 59 [12].

In our study, the incidence of SAH decreased by approximately 1% in 2020. The Korean health insurance system has contributed to this change through increased use of computed tomography angiography and magnetic resonance angiography during health screenings, a rise in brain computed tomography scans for patients presenting with headaches, educational initiatives on aneurysm treatment at academic conferences, and public health information campaigns. These factors have likely influenced the diagnosis and treatment of unruptured aneurysms. Clear causal relationships need to be further explored in future research.

Additionally, SAH tends to occur at a younger age compared to ICH or ischemic stroke. Therefore, incorporating cerebrovascular imaging into national health screenings for high-risk groups could aid in the early diagnosis and reduction of SAH incidence. This approach would require expanded insurance coverage and a nationwide consensus, necessitating changes in long-term national policies.

For ICH, it was associated with risk factors such as low cholesterol, high blood pressure, vascular malformations, and consistently associated with vascular malformations. Vascular malformations, a risk factor for ICH, are congenital malformations that can influence the early detection of ICH [5]. In the case of ischemic stroke, a total of 32259 patients (75.6%) were aged 60 or older, with three-quarters of all patients being elderly. A meta-analysis of ischemic stroke patients confirmed risk factors for elderly ischemic stroke such as thrombin generation, left ventricular mass, depressive symptoms, and low cognitive function [11]. These risk factors are direct causes of the high incidence of elderly ischemic stroke, and consideration should be given to revising guidelines to protect patients from ischemic stroke risk. In previous studies, it has been established that the elderly population is at a higher risk of vascular conditions such as hypertension and myocardial infarction. This elevated risk predisposes them to vascular-related issues like thrombus formation. Additionally, the increased prevalence of underlying conditions such as cardiovascular diseases, hypertension, and diabetes further contributes to the occurrence of acute ischemic stroke [1].

Among the demographic distributions of patients, the ratio of the proportion of assessments performed for severity (NIHSS/GCS) varied. Although the NIHSS can measure specific signs and severity of posterior circulation strokes, it was not effective in assessing chronic stroke [3]. Additionally, while the GCS predicted the mortality rate after acute ischemic stroke similarly to the NIHSS, it was more accurate in predicting poor neurological outcomes [13]. Although the NIHSS score was found to be correlated with outcomes in ischemic stroke, it was not correlated with outcomes in ICH patients [3]. Many grading scales have been proposed for SAH, but it is possible to make errors in dramatically improving or worsening treatment outcomes. To avoid such serious errors, it is suggested to base the GCS score itself, and applying GCS showed no errors mentioned above in SAH [16]. The severity assessment scale for stroke patients affected the prognosis prediction depending on the stroke subtype. This effect was shown as a result of the distribution of severity assessment by subtype.

Among domestic stroke patients, 27332 individuals (46.2%) received treatment at tertiary hospitals, with the proportion decreasing to 8695 (46.9%) from July 2016 to December 2016, 10026 (49.4%) from July 2018 to December 2018, and 8611 (42.3%) from October 2020 to March 2021. In contrast, patients treated at general hospitals increased to 31820 (53.8%), with proportions rising to 9835 (53.1%) from July 2016 to December 2016, 10263 (50.6%) from July 2018 to December 2018, and 11722 (57.7%) from October 2020 to March 2021. A similar trend of an increasing number of patients treated at general hospitals was observed in stroke sub-classifications, consistent with previous research indicating an increase in patients treated at general hospitals. In previous studies, the number of ischemic stroke patients treated at tertiary hospitals decreased from 42 institutions (20.9%) before 2015 to 25 institutions (18.1%) after 2015, while the number of general hospitals in this registry increased significantly from 140 institutions (69.7%) before 2015 to 213 institutions (85.9%) after 2015. These consistent findings suggest an improvement in accessibility to timely treatment for stroke patients at general hospitals, leading to an increase in the number of patients treated at general hospitals in the overall stroke classification analysis of this study. The high quality of medical accessibility in Korea is evidenced by 30676 patients (51.9%) arriving at hospitals within 5 hours of symptom onset (Fig. 1). Domestic medical accessibility played a crucial role in receiving treatment promptly for stroke, indicating that changes in guideline recommendations have influenced hospital facilities and patient care.

Fig. 1.

The proportion of hospital arrival time on acute stroke patients after symptom onset.

When examining the prognosis at discharge for acute stroke patients in South Korea by subtype, 3191 individuals (55.9%) with SAH had a favorable outcome upon discharge, while 26792 individuals (63.8%) with ischemic stroke had a favorable outcome upon discharge. More than half of the patients with SAH and ischemic stroke had a favorable outcome upon discharge. Among patients with ICH, 3598 individuals (34.5%) had a favorable outcome upon discharge, showing the lowest proportion of favorable outcomes among stroke subtypes. The age distribution of patients with hemorrhagic stroke, including ICH and SAH, did not differ significantly. For SAH, there are research findings indicating a high initial mortality rate but favorable outcomes for survivors. Conversely, for ICH, many survivors experience severe neurological sequelae due to the location of the hemorrhage affecting consciousness and basal ganglia or thalamus. These results appear to reflect the differences in prognosis at discharge between SAH and ICH.

This nationwide stroke patient analysis study also has several limitations. First, the absence of patient death data precluded tracking and observation based on stroke patient mortality. Generating long-term follow-up data using patient death dates could help identify trends in risk factors for acute stroke mortality, which could aid in informing stroke policy. Second, data on stroke patient brain imaging results were not included. Brain imaging results are an intuitive method for assessing severity in stroke patients. Although this study used NIHSS and GCS to assess stroke severity, incorporating imaging results for severity classification could provide more precise results.

This study is based on a population-based big data analysis that combines registry and HIRA data, with data representing a short-term period over 18 months. Due to national policy, data for 2016 and 2018 were collected only from July to December, while data for 2020 were collected from October to March. Consequently, we were unable to compare stroke incidence based on seasonal variations. This limitation should be addressed in future research. Furthermore, we were unable to conduct a meaningful analysis of the time difference from onset to hospital arrival between metropolitan and rural areas, the impact of hospital stay duration on patient outcomes at discharge, and the factor of COVID-19. These issues require further investigation and discussion in future research.

CONCLUSION

This study investigated the status of acute stroke patients and stroke sub-classifications based on the Korean Acute Stroke Acessment Registry. It provides an overview of the epidemiological characteristics of stroke in Korea, contributing to the improvement of stroke treatment guidelines in South Korea. Subsequent studies utilizing patient death data and brain imaging results should be conducted in the future.

Article information Continued