© M. Azeddou et al., Published by EDP Sciences, 2026

1 Introduction

Interventional radiology (IR) serves as a vital tool for assessing vascular anatomy. By offering high-resolution imaging of both arteries and veins, it plays a key role in diagnosing vascular conditions such as atherosclerosis and stenosis. Moreover, IR is instrumental in facilitating therapeutic procedures like angioplasty and stent deployment.

Despite these benefits, the technique often involves extended fluoroscopy times and elevated radiation exposure, posing potential health risks to patients and medical staff. This includes skin reactions like erythema, dermal atrophy, and ulceration and long-term radiation exposure concerns increasing risk of cancer (Vañó et al., 2017; Ferrari et al., 2020).

To help prevent these risks, ALARA is the fundamental radiation protection principle that applies to medical exposure, and diagnostic reference levels (DRLs) are one of the tools used to support the optimization process in medical settings (European Commission, 1999). DRL was introduced by ICRP in 1990, and defined as “thresholds for an investigation to optimize medical exposure during diagnostic and interventional procedures” (Vañó et al., 2017). DRL for interventional procedures are expressed in terms of four dosimetric metrics following the ICRP publication 135. These quantities include Dose Area Product (DAP), Air KERMA at the reference point (Kair) and Fluoroscopy Time (FT).

Despite its known limitations, such as not accounting for backscatter radiation, DAP remain the standard metric for setting DRLs both nationally and internationally. It serves as a surrogate indicator of stochastic risks, such as cancer. In contrast, (K_air) may offer more accurate insights into patient radiation safety, particularly regarding deterministic risks relating to skin (Vañó et al., 2017).

There is a notable lack of specific DRLs for IR procedures. While DRLs have been established for cardiologic procedures around the world (Siiskonen et al., 2018; Guenego et al., 2019; Ou-Saada et al., 2020; Ihn et al., 2021). In Morocco, there is still a lack of official bodies and policies overseeing national DRLs. Ou-Saada et al. (2020) proposed local DRLs for cardiology-related interventions, their study did not specifically address DRLs for other IR procedures (Ou-Saada et al., 2020). This gap is particularly concerning as the increasing use of these procedures could lead to higher radiation doses. Therefore, this study aims to establish local DRLs for the common IR procedures at the Mohammed VI University Hospital Center in Marrakech, Morocco.

2 Material and methods

A retrospective study was conducted in the medical Imaging department at the Mohammed VI University Hospital Center in Marrakech, between January and December 2024. The Cath-lab station included two rooms, one exploited by radiologists and the second belong to cardiology department. The angiography systems used were a Siemens Artis Zee, equipped with a flat panel detector. The study focused on IR guided procedures performed by radiologists and the cardiac procedures were excluded.

The sample included the most frequent IR procedures in 2024: Lower limb angiography (LLA), cerebral angiography (CA), Central venous access devices (CVAD), Pacemaker implantation (PM). Following ICRP recommendations, we manually extracted DAP, FT and K_air from the Digital Imaging and Communications in Medicine (DICOM) header (Vañó et al., 2017). A total of 220 of diagnostic procedures were included in this study. To further ensure accuracy, the X-ray system’s maintenance and calibration reports were reviewed, confirming that all quality control measures were up to date.

Data analysis was performed using Microsoft Excel 2017. The descriptive statistics were exploited to determine the mean, median, and 3^rd quartile values of dose distributions, including DAP, FT, and K_air. The local DRLs were defined as the 3^rd quartile. The derived local DRLs were compared to published studies DRLs.

3 Results

A total of 220 procedures were recorded at a single machine. Patient ages ranged between 20 and 80 yr. Table 1 presents four interventional procedures: LLA, CA, CVAD and PM. The key parameters reported include K_air in Gy, DAP in Gy·cm² and FT in minutes. The results are given in terms of mean, median, and 3rd quartile values.

For LLA procedures, the mean K_air was 0.30 Gy, with a median of 0.21 Gy and a 3rd quartile value of 0.47 Gy. The mean DAP was 81.38 Gy · cm², with a median of 59.34 Gy · cm² and a 3rd quartile of 107.98 Gy·cm². DAP and K_air exceed those of other procedures, which indicate that some cases involve significantly higher exposure. Comparing to the other procedures, FT remains relatively low, with a mean of 2.69 min and a median of 1.1 min.

For CA procedures, the mean K_air was 0.88 Gy, with a median of 0.78 Gy and a 3rd quartile of 1.08 Gy. The mean DAP was 196.61 Gy · cm², with a median of 184.93 Gy·cm² and a 3rd quartile of 232.18 Gy·cm². FT was significantly higher than other procedures, with a mean of 9.69 min, a median of 6.80 min, and a 3rd quartile of 11.40 min. These results indicate that CA procedures require prolonged fluoroscopic guidance, contributing to increased patient dose.

For CVAD procedures, radiation exposure was minimal. The mean K_air was 0.00871 Gy, with a median of 0.0046 Gy and a 3rd quartile of 0.0075 Gy. The mean DAP was 2.50 Gy · cm², with a median of 1.11 Gy · cm² and a 3rd quartile of 2.26 Gy·cm². FT was the shortest among all procedures, with a mean of 0.79 min, a median of 0.56 min, and a 3rd quartile of 1.15 min. These values confirm that CVAD involves low radiation exposure, likely due to the relatively simple nature of the procedure.

For PM procedures, the mean K_air was 0.10 Gy, with a median of 0.09 Gy and a 3rd quartile value of 0.13 Gy. The mean DAP was 20.74 Gy · cm², with a median of 16.87 Gy · cm² and a 3rd quartile of 27.10 Gy·cm². FT remains relatively high, with a mean of 6.95 min and a median of 5.90 min and a 3rd quartile of 7.98 min.

4 Discussion

The purpose of assessing Diagnostic Reference Levels (DRLs) was to pinpoint facilities where radiation protection practices may need further review to ensure optimization for patients and staff. This study may serve as a foundational step toward establishing national DRLs. Given the absence of national DRL guidelines in Morocco. Table 2 compares our local Diagnostic Reference Levels (DRLs) for four interventional procedures LLA, CA, CVAD, and PM based on DAP, FT, and Kair with values reported in previously published studies.

The findings indicate that CA procedures result in the highest radiation exposure across all three parameters (DAP, FT, and K_air). The high DAP values (196.61 Gy·cm²) in these procedures suggest extended fluoroscopic guidance and multiple image acquisitions to ensure precise catheter navigation. The prolonged FT (mean=9.69 min) was consistent with the complexity of CA interventions, which require meticulous positioning to minimize procedural risks. Consequently, the high K_air values (mean=0.88 Gy) indicate substantial radiation exposure to patients. However, CVAD procedures inherently demonstrate the lowest radiation exposure values.

The variability in dose metrics reflects the complexity of the procedure, which can differ depending on the clinical indication, even for the same type of procedure (Vañó et al., 2017). Additionally, this variability was influenced by the performance of the X-ray equipment and the patient’s size, which directly impact the exposure parameters (Crowhurst et al., 2019). The DAP differences, even with the same equipment, are largely influenced by operator factors. Since this study was conducted in an academic context, the critical aspect was the variability in radiologist experience. Distinctions between senior and junior practitioners impacted the outcomes, particularly in interventional settings (Crowhurst et al., 2019).

Comparison with previous studies highlights variations in radiation exposure across different healthcare systems and imaging protocols. For LLA procedures, the 3^rd quartile DAP in this study (107.98 Gy·cm²) was higher than values reported in France (75 Gy·cm²), Belgium (75 Gy·cm²), and Spain (73 Gy·cm²) but lower than those in Switzerland (210 Gy·cm²). Conversely, the FT in this study (2.73 min) is shorter than those reported in France (6 min) and the UK (5.9 min).

For CA procedures, the 3^rd quartile DAP (232.18 Gy·cm²) was largely higher than values from France (90 Gy·cm²) and the UK (61 Gy·cm²), but relatively higher than Switzerland (125 Gy·cm²). The FT (11.4 min) was within the range reported in France (11 min) and lower than Switzerland and Tristram et al. values. The large FT difference was within the Rizk et al. study (Tristram et al., 2022). The K_air (1.08 Gy) also exceeded values in France (0.63 Gy), Tristram et al. (0.34 Gy), and Rizk et al. (0.34 Gy) studies.

In CVAD procedures, the recorded 3^rd quartile DAP (2.26 Gy·cm²) was slightly higher than previous studies except from the UK (3 Gy·cm²), while FT (1.15 min) was comparable to international values. The K_air (0.0075 Gy) remains low, aligning with literature that supports the minimal radiation burden associated with peripheral angioplasty.

For PM procedures, the 3^rd quartile DAP (27.10 Gy·cm²) was higher than the UK (7 Gy·cm²) and other reported studies, while the FT (7.98 min) was comparable to international standards. The K_air (0.1268 Gy) was also within the range reported in previous studies.

The 3^rd quartile comparison revealed that the highest DAP difference was observed in PM procedures (+287%) and the lowest difference was found in CVAD procedures (-24.67%). When comparing FT, The highest difference was observed in LLA procedures (+82%). The lowest difference occurred in CA procedures (+3.64%). This trend was not surprising because of the absence of a radiation protection program based on DRL assessment. Studies conducted across multiple countries indicate that variations in regulatory frameworks can impact how DRLs are set and implemented (Siiskonen et al., 2018).

The results demonstrate the inconsistency between DAP and FT across almost all procedures. Similarly, the literature outlined a poor FT and strong DAP and revealed that the correlation between FT and dose metrics is very poor (Heilmaier et al., 2017). This was attributed to differences in imaging modes: acquisition mode increases the DAP and K_air without affecting FT, whereas fluoroscopic mode primarily contributes to FT without significantly increasing DAP and K_air (Tristram et al., 2022). The FT parameter is not a reliable measure of radiation damage and is less significant than DAP and K_air, making it inadequate for radiation protection purposes (Heilmaier et al., 2017; Tristram et al., 2022).

The underlying research methodology, such as sample sizes, varies across national DRLs and single facility studies, which accounts for the discrepancies. Our study reflected a small sample size for some procedures, compared to the national DRLs studies, which may skew our findings. Nonetheless, the ICRP states that such sample sizes are enough for implementing local DRLs (Vañó et al., 2017).

In accordance with ICRP Report 137 guidelines, this single-center study proposes the use of the median value as a local diagnostic reference level (DRL) to support the identification of potential areas for further optimization (Vañó et al., 2017; Tristram et al., 2022). However, this fact may limit the generalizability of the results to other institutions with different equipment, protocols, or operator practices. Analysis correlating dose level with procedure complexity and operator experience was not conducted, and this should be considered in future studies.

5 Conclusion

Local Diagnostic Reference Levels (DRLs) were determined for the common four IR procedures, based on Dose Area Product (DAP), air KERMA (K_air), and fluoroscopy time (FT). The resulting values surpassed those found in existing published DRL ranges, pointing to a clear need for enhanced dose optimization and refinement of clinical protocols. The findings underline the importance of completing and broadening this research to support the development of national reference levels for interventional radiology procedures in Morocco.

Funding

This research did not receive any specific funding.

Conflicts of interest

The authors declare that they have no conflict of interest.

Data availability statement

The research data associated with this article are included within the article.

Author contribution statement

Mina Azeddou, performed the research, analyzed data and wrote the paper.

Maroine Tahiri, performed the research and wrote the paper.

Mounir Mkimel, designed the study and writing-editing.

Bader Boutakioute, designed the study and writing-editing.

Najat Cherif Idrissi Elganouni, designed the study and writing-editing.

Ethics approval

Institutional Review Board approval was obtained.

Informed consent

This article does not contain any studies involving human subjects.

References

All Tables