112 R . Sindi et al .: Radioprotection 2024 , 59 ( 2 ), 111 – 116
indicative values of radiation dose for a given procedure based on audit data from a large number of patients at a local , regional or national level ( Alashban and Shubayr , 2022 ). DRLs represent the third quartile of the observed dose distribution for a given procedure based on audit data from numerous patients at a local , regional , or national level . DRLs are not dose limits , but rather benchmarks for comparison and optimization of patient exposure ( Damilakis et al ., 2023 ). The DRLs can be expressed in terms of dose indices such as CT dose index ( CTDI ) and dose length product ( DLP ), which reflect the output of the CT scanner ( Kanal et al ., 2017 ). DRLs can help to identify unusually high or low doses for a particular procedure , and to initiate actions to optimize the exposure parameters and the quality assurance program .
The DRL concept allows comparison of an individual facility ’ s typical radiation dose to the national benchmark for that examination . Regular nationwide surveys are required to update DRLs periodically to reflect evolving CT technology and practice ( Nam et al ., 2022 ; Abuzaid et al ., 2020b ). Many countries have implemented national DRLs ( NDRLs ) as a radiation protection tool to guide CT practice towards optimization ( Kumsa et al ., 2023 ; Rao et al ., 2023 ). Several studies have evaluated the implementation of NDRLs in different countries and regions and reported significant dose reduction and optimization for various CT examinations ( Korir et al ., 2016 ; Skovorodko et al ., 2022 ; Demb et al ., 2017 ; Wachabauer et al . 2020 ).
The Saudi Food and Drug Authority ( SFDA ) recently established national DRLs ( NDRLs ) for CT examinations commonly performed in Saudi Arabia . However , the impact of implementing DRLs on patient radiation exposure has not been extensively evaluated in Saudi Arabia . Therefore , the aim of this study is to investigate the impact of the implementation of NDRLs on patient radiation dose in terms of effective dose ( ED )& 9472 ; during chest CT examinations for adult patients in Saudi Arabia . Chest CT is one of the most frequently performed CT examinations , and it involves relatively high radiation doses , which make it a suitable candidate for dose optimization and audit .
2 Materials and methods
2.1 Study setting and data collection
Retrospectively , data were collected from four CT scanners ( SOMATOM Definition AS 64-slice , SOMATOM Definition flash dual-source 128-slice , Siemens , Munich , Germany ) at the Radiology Department in a major hospital in Jeddah , Saudi Arabia . The dataset comprises two-time frames : from March 2021 to February 2022 when NDRLs was not implemented yet ; and from March 2022 to February 2023 when NDRLs has been incorporated into clinical settings . Prior to commencing of the study , medical ethics approval was obtained from the Research Ethics Committee ( REC ) of King Fahd Armed Forces Hospital ( KFAFH ) ( REC 621 ). The collected data were anonymized , analyzed , and reported solely in an aggregate form .
2.2 Data extraction and parameters
The data was extracted from the Picture Archiving Communication System ( PACS ). The imaging parameters and dose metrics of CT chest imaging from adult patients were collected . The CT examinations included both contrasted and non-contrasted studies . Data comprised dose reports pertinent to the respective chest CT scan , including the scan protocol ( i . e ., kVp , mAs , scan length , DLP and CTDI vol ).
2.3 Calculation method for effective dose
To estimate the ED from Chest CT scans , we utilized the DLP , and a region-specific conversion factor known as the k- factor . The ED calculation is based on the formula
:
mSv
EDðmSvÞ ¼ DLPðmGy · cmÞ k � f actor mGy · cm
For our study , the k-factor for chest CT was sourced from the recently published data from a large registry / phantom library , taking into account scanner manufacturers ( Chu et al , 2023 ), which is different from previous k-factors that were derived from theoretical models rather than real patient data . As stated by Chu et al ., “ the new coefficients offer reasonably reliable values for estimating effective dose .” The k-factor for chest CT for Siemens scanner is 0.040 mSv /( mGy · cm ) ( Chu et al ., 2023 ). It should be noted that value of 0.040 is about 1.8 and 2.6 times the previous values of 0.020 , 0.014 , and 0.0146 from previous published k-factors ( Huda et al ., 2008 ; Deak et al ., 2010 ; Shrimpton et al ., 2006 ). In addition , the ED is not intended for dose to an individual ; intended for populations .
2.4 Reference levels
Based on the NDRLs for chest CT , the reference levels for CTDIvol and DLP are set at 12 ( mGy ) and 430 ( mGy · cm ), respectively . These levels act as benchmarks to assess pass / fail outcomes in imaging procedures . For this study , a procedure recording a CTDIvol over 12 or a DLP over 430 would be classified as a failure . Conversely , if both CTDIvol and DLP values are at or below these thresholds , the procedure is deemed a pass . Applying these standards to the provided data , any case with a CTDIvol exceeding 12 is labeled as “ Fail CTDIvol ,” while those surpassing a DLP of 430 are marked as “ Fail DLP .” Cases exceeding both limits are categorized as “ Fail CTDIvol and DLP ,” and those within the thresholds for both measures are classified as “ Pass .”
2.5 Statistical analysis
Statistical analyses were performed using SPSS Statistics software ( SPSS version 26 , IBM , USA ). Descriptive and inferential statistics were conducted . Percentage change was conducted for prior to and after the implementation of NDRLs for radiation parameters . Mann – Whitney U test was used to compare the ED prior to and after the implementation of NDRLs . A p value of 0.05 was set for significance .
3 Results
In our study , we analyzed data from a total of 1,991 chest CT scans . 429 scans ( 21.5 %) were performed with contrast , while 1,562 scans ( 78.5 %) were conducted without contrast . For collected data before the implementation of NDRLs ,