PHWR Public Health Weekly Report

Cardiovascular diseases in the Republic of Korea (ROK) are among the leading causes of death, and the use of image-based interventional procedures for the diagnosis and treatment of these diseases has been steadily increasing [1,2]. Angiography and interventional procedures can be performed with relatively low invasiveness compared with conventional invasive treatments and therefore constitute an indispensable component of medical diagnosis and treatment [3]. However, because angiographic and interventional procedures are performed based on high-resolution imaging and prolonged fluoroscopy, the radiation dose delivered to patients is relatively higher than that received during general imaging examinations [2,3]. According to the 2023 annual report on medical radiation exposure of the population published by the Korea Disease Control and Prevention Agency (KDCA), the use of medical radiation among the population has been steadily increasing, with angiographic examinations increasing by approximately 21.3% in 2023 compared with 2020. As of 2023, angiographic examinations accounted for only 0.2% of the total use of medical radiation, but they contributed to 2.3% of total patient radiation exposure, ranking third among medical examinations in terms of radiation dose [4]. The effective dose per capita from angiographic examinations is 0.07 mSv. Among these procedures, coronary angiography (CAG) accounted for the largest proportion (39.3%) of all angiographic examinations and showed a relatively high contribution to radiation exposure (26.4%) [2,4].

Six international organizations, including the World Health Organization (WHO) and the International Atomic Energy Agency (IAEA), jointly recommended the management of patient radiation dose through the international basic safety standards No. 115 in 1996 [5]. In addition, the International Commission on Radiological Protection (ICRP) recommends the establishment and use of diagnostic reference levels (DRLs) to optimize patient radiation exposure, and dose management and optimization based on these levels have become important components of healthcare systems worldwide [6,7]. DRLs are defined as the third quartile (75th percentile) of the radiation dose distribution obtained by surveying the radiation doses used in diagnostic imaging examinations at individual medical institutions nationwide [6-9].

Developed countries, including those in Europe, Japan, and the United States, promote radiation dose management and quality improvement by periodically updating national DRLs. In the ROK, national DRLs were first established and announced by the Ministry of Food and Drug Safety in 2008. Following the transfer of responsibility in 2013, the KDCA has been presenting national DRLs for various types of medical radiation examinations every 3–5 years [8,9]. To date, national DRLs have primarily focused on major examinations such as general radiography and computed tomography (CT), while DRLs for interventional radiology procedures, particularly cardiovascular interventions, have not yet been sufficiently established [8-10].

In the absence of DRLs for high-dose procedures, objective criteria for variations in radiation dose between medical institutions or operators cannot be established, which might lead to unnecessary increases in patient radiation exposure. This situation may contradict the radiation protection principle of dose optimization [6-8]. Therefore, the establishment of objective standards for interventional procedures is urgently needed, and the development of DRLs based on real clinical data is essential. Accordingly, this study collected and analyzed radiation dose data from CAG and interventional procedures performed at major medical institutions in the ROK and aimed to propose procedure-specific DRLs, thereby providing baseline data for the establishment of future clinical and policy standards.

1. Study Population

1) Examinations and procedures

According to ICRP publication 135, when determining which examinations or procedures should have DRLs established, priority should be given to procedures that are commonly performed in the region, particularly those performed most frequently or those associated with the highest patient radiation exposure [6,7]. The main variables to be recorded should be quantities that can be easily evaluated and, if possible, should be radiation dose metrics that can be directly measured during the examination or obtained from the imaging equipment [6,7]. Accordingly, this study selected seven examinations that are most frequently performed in the field of CAG and interventional procedures. In this study, acute myocardial infarction (AMI) and chronic total occlusion (CTO) were treated not simply as diagnostic labels but also as representative clinical categories wherein lesion characteristics might increase procedural difficulty and the amount of fluoroscopy and image acquisition. Therefore, cases wherein interventional procedures for these lesions were actually performed were defined as separate procedural groups. The selected examinations and procedures were as follows: (1) CAG; (2) Coronary angiography and percutaneous coronary intervention (CAG+PCI); (3) Coronary angiography and percutaneous transluminal coronary angioplasty (CAG+PTCA); (4) Coronary angiography and spasm provocation test (CAG+SPASM); (5) AMI procedures; (6) CTO procedures; (7) PCI.

2) Selection of participating medical institutions

According to ICRP publication 135, sample surveys conducted to establish DRLs should collect sufficient data to obtain a representative patient sample, and it is recommended that at least 20 patients be collected for a specific examination from a single facility. Furthermore, it is stated that when a sample survey is conducted by randomly selecting a subset of facilities from all facilities, results from 20–30 facilities are sufficient to estimate national DRLs [6,7].

The status of accredited medical institutions capable of performing CAG and interventional procedures was identified through the database of the Korean Society of Interventional Cardiology. Subsequently, the institutions were contacted to confirm their willingness to participate in the study. To facilitate efficient data collection, 20 medical institutions equipped with systems capable of generating digital imaging and communications in medicine (DICOM) radiation dose structured reports (RDSR) and procedure reports after the completion of procedures were selected, and radiation dose–related information was collected for the establishment of DRLs.

2. Data Collection and Analysis

According to ICRP publication 135, several metrics may be used for establishing DRLs for angiography and interventional procedures, including dose area product (DAP), cumulative air kerma, fluoroscopy time (FT), and the number of images acquired through cine imaging or digital subtraction angiography [6]. In this study, DICOM RDSR files and procedure reports generated from angiography systems installed at participating medical institutions were used for data collection. The RDSR is a structured dose report defined according to the DICOM standard and quantitatively includes various exposure-related information, such as device type, patient sex, age, height, weight, examination name, cumulative DAP for the entire examination, DAP for each image, and FT [11]. Therefore, the RDSR is suitable for use as an indicator for estimating DRLs. Depending on the network environment of each institution, RDSR files can be transmitted to an external server through linkage with the imaging equipment server or the picture archiving and communication system [11].

Accordingly, for data collection, the research team conducted on-site inspections of the participating medical institutions. Depending on whether network access was available, institutions were allowed to provide data using one of two methods: real-time transmission of data or local storage followed by subsequent collection. In addition, for institutions where opening the DICOM network was not possible due to institutional constraints, procedure reports generated after the completion of procedures were collected in parallel. For procedure report data collection, an Excel (Microsoft) data form was used to receive the entered information, retrieve the completed forms, and upload them to the study’s internal data collection server for analysis. Among the total of 20 participating medical institutions, data from 14 institutions were obtained through an RDSR-based real-time transmission method, while the remaining six institutions used a procedure report-based data collection method due to network limitations. A schematic diagram of the data collection process is shown in Figure 1.

Figure 1. Schematic diagram of radiation dose data collection for coronary angiography and interventional procedures
DICOM=digital imaging and communications in medicine; XA=X-ray angiography; RDSR=radiation dose structured reports; SW=software; DRLs=diagnostic reference levels.

Data collection was conducted from the second quarter to the fourth quarter of 2024. Considering the consistency of the data and the feasibility of data collection across participating institutions for each examination type, cumulative DAP (Gy‧cm²) and FT (second), which were available from all institutions, were selected as the final analytical indicators and collected. Cumulative DAP and FT are procedure-based indicators calculated at the completion of a procedure and reflect the level of radiation exposure delivered to the patient during the procedure. The collected raw data were aggregated on the study’s internal server and organized into an integrated database. To prevent the leakage of sensitive personal information, technical security measures were implemented, and all institutional identifiers and patient information were removed during preprocessing to ensure that the data were fully anonymized.

To analyze the collected data, descriptive statistical analysis was performed for each variable. For each examination type, the mean, quartiles (Q1 and Q3), and median values were calculated for cumulative DAP and FT. DRLs were defined based on the third quartile of the calculated dose values. In addition, the first quartile (25th percentile) was defined as the lower DRLs, and the median was defined as the target dose.

A total of 1,980 cases were collected during the second to fourth quarters of 2024 for the seven types of CAG and interventional procedures included in this study. The unit of analysis in this study was a single procedure, and only valid cases wherein cumulative DAP and FT were available were included. All participating institutions satisfied the minimum sample size criterion (at least 20 cases per institution). Based on these data, the 75th percentile values were calculated using descriptive statistics for each procedure type. The results showed that DAP values were 18.68, 63.40, 53.89, 25.44, 58.52, 106.83, and 49.94 Gy‧cm² for CAG, CAG+PCI, CAG+PTCA, CAG+SPASM, AMI, CTO, and PCI, respectively. The 25th percentile, median, and mean values of DAP are presented in Table 1.

Table 1 Distribution of dose area product for CAG and interventional procedures

Procedure name	25th percentile	Median	75th percentile	Average
CAG	6.22	10.17	18.68	14.92
CAG+PCI	24.20	39.13	63.40	48.78
CAG+PTCA	25.20	33.84	53.89	42.89
CAG+SPASM	10.89	17.90	25.44	18.68
AMI	29.13	43.10	58.52	56.38
CTO	48.98	73.56	106.83	78.95
PCI	11.74	21.82	49.94	34.13

Unit: Gy‧cm². CAG=coronary angiography; PCI=percutaneous coronary intervention; PTCA=percutaneous transluminal coronary angioplasty; SPASM=spasm provocation test; AMI=acute myocardial infarction; CTO=chronic total occlusion..

The 75th percentile values of FT were 440.00, 1,201.50, 1,284.65, 341.26, 947.64, 2,819.00, and 1,479.50 seconds for CAG, CAG+PCI, CAG+PTCA, CAG+SPASM, AMI, CTO, and PCI, respectively. The 25th percentile, median, and mean values of FT are presented in Table 2.

Table 2 Distribution of fluoroscopy time for CAG and interventional procedures

Procedure name	25th percentile	Median	75th percentile	Average
CAG	135.00	224.53	440.00	395.84
CAG+PCI	520.00	817.50	1,201.50	938.28
CAG+PTCA	469.72	673.84	1,284.65	1,091.80
CAG+SPASM	174.80	232.12	341.26	298.52
AMI	439.00	671.41	947.64	874.69
CTO	1,203.18	2,143.22	2,819.00	2,099.02
PCI	555.99	982.00	1,479.50	1,192.61

Unit: sec. CAG=coronary angiography; PCI=percutaneous coronary intervention; PTCA=percutaneous transluminal coronary angioplasty; SPASM=spasm provocation test; AMI=acute myocardial infarction; CTO=chronic total occlusion..

The 75th percentile values of cumulative DAP and FT derived from these data were established as the national DRLs for the seven CAG and interventional procedures.

National DRLs for angiography and interventional procedures in the field of radiology in the ROK were established twice, in 2012 and 2019 [12]. However, for cardiovascular procedures, available information had been limited to radiation dose data reported by individual researchers in the ROK [10]. Accordingly, this study is meaningful as a policy-oriented study that collected radiation dose data from multiple medical institutions and analyzed them by procedure type in order to estimate the first national DRLs for CAG and interventional procedures in the ROK.

In this study, multicenter clinical data from accredited medical institutions were analyzed, thereby overcoming the limitations of single-institution studies and increasing the representativeness and external validity of the results. In addition, while utilizing quantitative dose indicators derived from DICOM RDSR, cumulative DAP and FT, which could be commonly collected from all institutions, were selected as analytical variables in order to ensure comparability among institutions. By establishing the third quartile of the dose distribution as the national DRLs for seven procedures frequently performed in clinical practice, this study presents a more standardized, data-based, and consistent estimation framework compared with previous national DRLs studies.

A comparison of the calculated DRLs by procedure type showed that CTO exhibited the highest DAP (106.83 Gy‧cm²) and the longest FT (2,819.00 seconds). This finding reflected the characteristics of CTO lesions, which are chronically occluded and often involve long lesions or severe calcification. These factors increase procedural difficulty and are associated with complex procedural processes and a tendency toward prolonged FT [3]. These clinical characteristics lead to prolonged fluoroscopy and repeated image acquisition, ultimately resulting in increased patient radiation exposure. In contrast, because CAG is mainly centered on simple image acquisition for diagnostic purposes, it showed the lowest DAP at 18.68 Gy‧cm², and its FT was also confirmed to be the second shortest among the procedures at 440.00 seconds.

In the case of combined procedures such as CAG+PCI, CAG+PTCA, and CAG+SPASM, radiation exposure was found to be relatively higher than that of purely diagnostic procedures. In particular, the DAP of CAG+PCI was approximately 3.4 times higher than that of CAG, with three times longer FT, indicating approximately 12.7 minutes longer radiation exposure time. This finding might be attributed to the structural characteristic that the procedural process becomes more complex as interventional treatment is performed consecutively following diagnostic angiography [3].

When the DRLs derived in this study for CAG and PCI were compared with values aggregated by continent [13], the DRLs were found to be relatively lower (Figures 2, 3). This finding suggested that medical institutions in the ROK are actively adopting advanced equipment and radiation dose–reduction protocols. A previous study by Lee et al. [14] comprehensively analyzed the status of DRLs establishment for interventional procedures in the ROK and other major countries and reported that radiation dose levels vary substantially across procedures and that differences exist among countries in terms of the criteria and numerical values used for DRLs establishment. In particular, interventional procedure DRLs in the ROK were generally lower than those in the United Kingdom, Germany, and the United States for most procedure categories, which might be attributed to radiation dose optimization efforts and the use of advanced equipment in some leading medical institutions [14].

Figure 2. Diagnostic reference levels of CAG published for different regions
When several works were compiled, the boxes represent the quartiles and whiskers show the 5% and 95% percentile. CAG=coronary angiography.

Figure 3. Diagnostic reference levels of PCIs published for different regions
When several works were compiled, the boxes represent the quartiles and whiskers show the 5% and 95% percentiles. PCI=percutaneous coronary intervention.

The establishment of national DRLs is not limited to the simple presentation of numerical values but can also support various applications in clinical practice. Medical institutions can perform self-assessment of radiation dose levels using DRLs as a benchmark and can use them as reference materials for equipment performance evaluation and the development of operational guidelines. In addition, DRLs can be used to improve operators’ awareness of radiation dose and as educational materials, thereby contributing to the establishment of a radiation safety culture. Furthermore, at the national level, DRLs can support comparative analyses among medical institutions and the development of feedback systems, and in the mid-term to long-term, they can serve as baseline data for the establishment of radiation dose optimization policies.

For the effective implementation and operation of DRLs, continuous and systematic data collection should be conducted in parallel. In this study, the validity of DRLs establishment was confirmed through DICOM RDSR-based data, suggesting the potential for the future introduction of an automated radiation dose monitoring system. However, during the pilot operation process, some medical institutions experienced practical difficulties in applying computerized workflows, and the specific characteristics of procedures, such as changes in orders during the procedure and the occurrence of emergency procedures, acted as factors that hindered system integration.

In addition, the fact that data collection was concentrated within a specific period should also be taken into consideration when interpreting the results. The data collection period of this study partially overlapped with a period during which clinical care and procedural operations were subject to change due to collective actions within the Korean medical community, and thus, the number of procedures performed at some institutions may have decreased compared with typical years. Therefore, the DRLs derived in this study might reflect the clinical operational conditions during that period, and future re-evaluation and updating based on accumulated data over a longer period are needed.

In this study, because the focus was placed on deriving representative values for national DRLs using RDSR-based multicenter real-world clinical data, detailed stratified analyses according to device characteristics, patient characteristics (sex, age, height, and weight), and operator-related factors were not performed. However, ICRP publication 135 states that appropriate DRLs quantities can be recorded using automated methods and that, in sample surveys where a sufficient number of patients (>100) is secured at each facility, restrictions related to patient factors, such as body weight, may be relaxed [6]. In accordance with these recommendations, this study adopted an approach to derive representative values based on large-scale multicenter clinical data. In addition, given the characteristics of angiography and interventional procedures, radiation dose variation may occur depending on equipment configuration, procedural difficulty, operator proficiency, and lesion characteristics, and the results of this study are subject to some limitations in interpretation in that stratified analyses according to these factors were not performed. In the future, additional detailed analyses reflecting patient characteristics, equipment characteristics, and procedural difficulty are warranted.

In conclusion, this study was significant as it presented national DRLs for CAG and interventional procedures based on multicenter real-world clinical data and proposed their potential utility and the importance of their clinical application. In the future, additional efforts will be required, including the establishment of a real-time feedback system based on national DRLs, expansion to comparative evaluation indicators among medical institutions, and the establishment of more detailed criteria for high-difficulty procedures. Through such efforts, it will be possible to contribute to the creation of a patient-centered and safe radiation-use environment and to the realization of radiation dose optimization.

Ethics Statement: The study protocol was approved by the Institutional Review Board (IRB) of Daegu Health College (DHCIRB-2024-03-001).

Funding Source: This work was supported by the Korea Disease Control and Prevention Agency (2024-10-003).

Acknowledgments: The authors would like to thank all the researchers who contributed to this study: Kukjin Jeon (Bundang Cha hospital).

Conflict of Interest: The authors have no conflicts of interest to declare.

Author Contributions: Conceptualization: JSK. Data curation: DHK, BRC. Formal analysis: DHK, BRC. Funding acquisition: JSK. Methodology: YJM, JWG, JHW. Project administration: JSK. Supervision: JSK, HMP. Validation: HMP. Visualization: DHK. Writing – original draft: DHK. Writing – review & editing: HMP, JSK.