114 A. El Khatib et al.: Radioprotection 2026, 61( 2), 113 – 120
States( Ataç et al., 2015; McDermott et al., 2019). Studies emphasize the importance of optimizing radiation doses in CT examinations to mitigate potential risks, especially considering the association between ionizing radiation and the development of neoplasia( Power et al., 2016). Despite ongoing debates about the cancer risks associated with CT scans, efforts towards dose reduction and optimization remain paramount in ensuring patient well-being and minimizing unnecessary radiation exposure( McCollough et al., 2015). To mitigate these risks, it is crucial to promote optimization processes aligned with the As Low As Reasonably Achievable( ALARA) principle( Kalender, 2014). One of the key optimization tools is the implementation of Diagnostic Reference Levels( DRLs), which serve as investigative thresholds to discern radiation doses that deviate significantly from the norm( Nenot et al., 2009). However, there is a notable gap in DRLs for the CTA protocols in Morocco( Mohiy et al., 2012).
DRLs, as delineated by the International Commission on Radiological Protection( ICRP) in 1996, are established based on the CT dose index volume( CTDI vol) and dose length product( DLP)( ICRP, 2001). In 2000, the mandatory incorporation of DRLs was instituted at the European level, with successive surveys conducted across various European nations, including Italy( Palorini et al., 2014), the United Kingdom( Shrimpton, 2003), Ireland( Foley et al., 2012), and Switzerland( Treier et al., 2010). The UK study observed a notable 50 % reduction in average dose from 1985 to 2000, attributed in part to the utilization of DRLs( Shrimpton, 2003; Wallace, 2010). DRLs are typically derived from the 50 th or 75 th percentile values of CTDI vol and DLP, with the ICRP advocating for their establishment through patient dose assessments at local, regional, or national levels( ICRP, 2017). To our knowledge, no study has been conducted to determine DRLs for CTA in Morocco.
The primary objective of this study was to establish local DRLs and assess radiation doses for CTA examinations on adult patients. The investigation was conducted at two public and two private hospitals across Morocco. Using the 75th percentile of CTDIvol and DLP distributions, local DRLs were determined and compared with international DRL values.
2 Materials and methods
2.1 Data collection
Data collection was performed retrospectively from the Picture Archiving and Communication System( PACS) for each hospital, on a cohort of 2283 adult participants( 951 male and 1332 female), with a mean age of 61 yr( 16 to 113 yr). The study included a minimum of 50 CT exams per hospital, covering at least three types of CTA examinations. Data was collected from two large public medical centers and two private clinics, and was categorized by CTA examination type and the CT systems used to analyze the frequency of the examinations. Patient confidentiality and anonymity were respected. Patient dosimetry was based on the CTDIvol and DLP. The 75th percentile values from their distributions were proposed as LDRLs and compared with international DRL values.
2.2 CTA protocols
The CT scanners used in this study were distributed as follows: Hospital A utilized a General Electric( GE) Healthcare Optima CT 540( 16 detectors), and Hospital B employed a GE EVO-REVO( 64 detectors). Hospital C used a Philips Brilliance scanner( 16 detectors), while Hospital D operated a Siemens SOMATOM go now( 16 detectors). Most of the examined protocols are equipped with tube current modulation software, except for the brain CTA examinations.
2.3 Statistical analysis
Statistical analysis was performed using IBM SPSS Statistics v. 20. Descriptive statistics( mean, median, range) were calculated for the demographic and dosimetric data( CTDIvol and DLP) from each hospital. Local DRLs were determined from the 75th percentile of the CTDIvol and DLP. An Analysis of Variance( ANOVA) was used to assess dose differences for identical CTA examinations across the four CT systems.
3 Results
Dose indicator data, CTDIvol and DLP, are presented in Tables 2 and 3, respectively, which include mean values and local DRLs for the four hospitals, with LDRLs also compared to international DRL values in Table 4.
An Analysis of Variance( ANOVA) revealed a highly significant difference( p < 0.001) in CTDI vol between hospitals for all vascular examinations analyzed. The mean CTDI vol values for various CTA examinations are presented in Table 2. The variations in CTDI vol values were as follows For Brain CTA, a 42.12 % difference was observed between institutions. In Carotid CTA, Hospital B recorded the highest value of 17.29 mGy, with discrepancies of 52.04 %, 78.64 %, and 110.69 % compared to Hospitals A, C, and D, respectively. For Pulmonary Arteries CTA, Hospital C exhibited the highest value of 11.20 mGy, with differences of 40.34 %, 108.12 %, and 56.59 % compared to Hospitals B, D, and A, respectively. In Abdominal Aorta CTA, CTDI vol values were similar between centers B and C, however Hospital A registered the highest value, with differences of 28.84 % and 43.18 % compared to Centers B and C, respectively. Total Aorta CTA revealed a value of 7.53 mGy at Hospital A, with a difference of 40.55 % compared to center C. Lastly, Lower Limb CTA showed the highest value at Hospital B, surpassing Hospitals C, A, and D with differences of 48.53 %, 75.91 %, and 84.73 %, respectively( as shown in Tab. 2)
An analysis of DLP values across various CTA examinations and hospitals using the analysis of Variance( ANOVA) showed significant differences for most comparisons. Notably, no significant difference was found for Brain CTA( p = 0.823) and the A-D hospital pair for lower limb CTA( p = 0.219). Conversely, significant variations( p < 0.05) were observed in mean DLP values for Carotid, Pulmonary Arteries, Abdominal Aorta, Total Aorta, and several pairs of hospitals for lower limb CTA( as shown in Tab. 3). For Brain CTA, a difference of 10.32 % was observed between Hospitals A and B. In Carotid