Monitoring Pediatric Head CT Scan Dose Levels: A Retrospective Study of Diagnostic Reference Levels in a Single Hospital in Abu Dhabi, UAE

Abstract

Diagnostic reference levels (DRLs) are critical operational standards for monitoring radiological imaging parameters and ensuring patient safety in hospitals. A regular review of DRLs is necessary to optimize scanners and protocol parameters. This retrospective study collected data on the CT dose index volume and dose–length product for 250 children who underwent head CT scans in the region of Abu Dhabi, United Arab Emirates. Descriptive statistics were used to present median, range, and interquartile range values for each pediatric age group, based on region and protocol. The 75th percentile was used as the reference point for local DRLs. Proposed values of DRLs were suggested per age groups; 304 mGy∙cm (children < 1 year), 385 mGy∙cm (children ≥ 1 year to <5 years), 441 mGy∙cm (children ≥ 5 years to <10 years), and 568 mGy∙cm (patients ≥ 10 years to <16 years. A comparison of the local DRLs with previously established ones was carried out, and differences were discussed. To the best of our knowledge, this is the first study on pediatric head CT scans in this region. We believe the results can be used as a baseline for establishing local DRLs in the Emirate of Abu Dhabi and national DRLs in the United Arab Emirates.

1. Introduction

The role of diagnostic imaging in the diagnosis and treatment of disease is very important. Different techniques and tools are used in radiology in order to provide information and visualize the inner body of a patient [1,2]. Nevertheless, the human body may suffer harm from exposure to the radiation used in X-rays, specifically in children, due to their underdeveloped organs and tissues. For the last twenty years, several international standards have supported radiation dose optimization in medical exposure, especially in computed tomography (CT) scans. These include the International Atomic Energy Agency (IAEA), the International Basic Safety Standards (BSS), the European Basic Safety Standards, as well as the International Commission on Radiological Protection [2,3,4,5]. The primary goal of these standards is to reduce the risks associated with radiation exposure while ensuring that the necessary diagnostic information is obtained. Therefore, the use of appropriate imaging protocols and dose-reduction techniques is essential to minimize radiation exposure in pediatric patients.

Due to the technological advances in CT scanners, they are considered essential diagnostic tools by many medical practitioners [6]. CT scanning is an essential medical imaging technique that has significantly contributed to the diagnosis and treatment of various medical conditions. CT scans provide detailed images of internal body structures, making it possible to identify abnormalities and injuries that may not be visible through other diagnostic tests. They are particularly useful in emergency situations, where quick diagnosis and treatment are crucial for saving lives.

CT scanning is a popular diagnostic tool used to identify various medical conditions such as cancer, trauma, cardiovascular disease, and infectious diseases [7,8,9]. It plays a key role in diagnosing and staging cancer and evaluating trauma patients. However, CT scanning involves a higher radiation dose than conventional radiography, which poses a significant public health concern. To minimize radiation exposure, it is important to monitor radiation exposure and implement appropriate safety measures, especially during pediatric CT imaging [7,10]. The establishment of a diagnostic reference level (DRL) is an essential optimization strategy for ensuring that the benefits of CT scans outweigh their risks [11]. In 1996, the International Commission on Radiological Protection (ICRP) introduced diagnostic reference levels (DRLs) as tools to optimize and monitor the radiation levels to which patients are exposed during medical imaging procedures [7]. DRLs play a crucial role in managing radiation doses in medical imaging procedures, particularly in pediatric CT imaging. By establishing and following DRLs, medical facilities can help to ensure that radiation doses are kept as low as reasonably achievable while still providing the necessary diagnostic information, ultimately protecting the health and safety of patients. In addition, the official guidelines recently published by the European Commission report the methodology to be followed to establish DRLs exclusively for pediatric patients [12]. According to the guidelines’ standards, the national DRL should be based on patient dose surveys involving samples from a large number of institutions within a country. Similar to national DRLs, the establishment of local DRLs is also based on patient dose surveys involving representative samples within a region or facility. They are based on the 3rd quartile or the 75th percentile of the median values of the distribution of patient doses. Following these guidelines, different DRLs have been established in several countries or regions. This variety of DRLs is mainly due to variations in equipment, procedural protocols, and the size and weight of the patients in the corresponding countries or regions [13]. Furthermore, DRLs are set using standard CT parameters, including the CT dose index volume (CTDIvol) and dose–length product (DLP).

The CTDIvol is a measure of the radiation dose delivered to a patient during a CT scan. It takes into account both the radiation intensity and the scan range and is expressed in units of milligray (mGy) [7]. Milligray is the standard metric used in CT dose monitoring and is used to estimate the dose delivered to a patient during a scan. The CTDIvol value is displayed on the CT scanner and is recorded in the patient’s medical records.

The DLP is a measure of the total radiation dose delivered to the patient during the entire CT scan and is typically reported in units of mGy·cm. The DLP is calculated by multiplying the CTDIvol by the length of the scan, and it provides an estimate of the total radiation exposure to the patient [7]. DRLs for CT scans are typically established using a combination of CTDIvol and DLP and are based on the radiation doses delivered to a representative sample of patients undergoing the same imaging procedure. The establishment of DRLs helps to ensure that radiation doses are kept as low as reasonably achievable while still providing the necessary diagnostic information. The use of both CTDIvol and DLP in the establishment of DRLs is important because it provides a comprehensive assessment of the radiation dose delivered to the patient during CT scanning [12]. While the CTDIvol provides information on the radiation dose delivered per unit length of the scan, the DLP provides an estimate of the total radiation exposure to the patient, taking into account the length of the scan.

In this context, this study is the first attempt, in the region, to summarize the results of a pediatric head CT survey in the Emirate of Abu Dhabi, United Arab Emirates (UAE). The work aims to provide local DRLs for the Emirate of Abu Dhabi, as well as contribute to national efforts to establish national DRLs for the country.

2. Materials and Methods

The current study has been approved by the Health Service Company (SEHA), the research committee of Abu Dhabi, UAE, with the approval reference number SEHA-IRB-022 dated 27 April 2021. The SEHA is the official institution that owns and operates public hospitals and clinics within the Emirate of Abu Dhabi, the largest province in the United Arab Emirates (UAE) [14]. In addition, the Federal Authority for Nuclear Regulation (FANR) performs periodic checks on the readings of all CT systems within the country [15].

Data on pediatric cases from January 2020 to December 2020 were obtained from the Sheikh Shakboot Medical City (SSMC), the largest hospital in the region, which is operated by the Mayo Clinic. The SSMC includes 732 medical beds [14]. The data were extracted from the Picture Archiving Communication System (PACS). SEHA’s PACS system specialists pulled 250 pediatric head CT scans at random. The files included patient protocol files and dose reports related to each head CT scan procedure. In addition to the date of the CT scan and the type of CT machine, the protocol files and dose reports contain information regarding the patient’s age, the tube voltage, tube current–time product, computed tomography dose index volume (CTDIvol), and dose–length product (DLP). The CTDIvol for head CT examinations was based on a 16 cm diameter phantom. The DLP is calculated by multiplying the CTDIvol by the length of the scanned volume and is reported in units of milligray centimeters (mGy·cm). Patient consent requirements are waived for retrospective data collection. All the collected medical image files were processed and preserved in the DICOM format using MicroDicom viewer software. A Python code was written to extract the required data from the 250 DICOM files. Data subjects were excluded if they had multiple exams elsewhere in the body or insufficient data from dose reports. The retrieved data were stored in a Microsoft Excel spreadsheet and divided into 4 groups based on their ages: Group 1 (children ≤ 1 year of age), Group 2 (children > 1 year to ≤5 years), Group 3 (children > 5 years ≤10 years), and Group 4 (patients > 10 years ≤16 years).

CTDIvol and DLP data were analyzed according to the minimum and maximum values, and the difference between them (i.e., the range). To ensure the accuracy of the analysis, a number of outlier data points were identified and excluded. Specifically, data points with very low values were excluded to ensure the robustness of the analysis. Furthermore, the values of the 1st quartile (25th percentile), 2nd quartile (median, 50th percentile), and 3rd quartile (75th percentile) were calculated. Finally, the 75th percentile was used to make a comparison with the established DRLs from other published studies. In order to find relevant original research studies, a thorough search of the literature was performed on Web of Science, Scopus Medline, ProQuest Health, Medical Complete, PubMed, and ScienceDirect to find pediatric CT studies that have established DRLs. The search terms used were “diagnostic reference levels”, “pediatrics”, “computed tomography”, “CT”, and “radiation dose”. The search was restricted to research articles published in the English language between 2015 and 2020. We intentionally focused on the most recent years’ articles in order to have access to the most recently established DRLs. Articles were excluded if they (i) were conference abstracts, (ii) did not propose a pediatric DRL for computed tomography, and (iii) were phantom studies.

3. Results and Discussion

The scanner installed in the SSMC hospital is manufactured by Siemens (Siemens Healthineers, SOMATON machine model, Erlangen, Germany). This scanner is equipped with dedicated pediatric CT imaging including 70 KV scan modes and specific CARE Dose4D, a Siemens technology for automated exposure control. The scan acquisition parameters are presented in

Table 1. Exposure parameters (tube voltage and tube current–time product) for the age groups considered in this study.

	Exposure Parameters
Age Group	Tube Voltage (kVp)	Tube Current–Time Product (mAs)
<1 year	80–100	110–297
≥1 y <5 y	80–120	67–366
≥5 y <10 y	80–120	77–581
≥10 y <16 y	80–120	63–534

. The patient tube voltage for the children age groups considered in this study was between 80 kVp and 100 kVp for patients younger than 1 year. For the other age groups, this parameter was between 80 kVp and 120 kVp. The tube current–time product ranged from 67 mAs to as high as 534 mAs. The lower limit value of the tube current–time product for the smallest age group was higher than that of the other groups. Despite the higher lower limit value of the tube current–time product for the youngest age group, the overall dose received by the patients was still within the safety standards, as we will see later. This suggests that in the CT scanner used in this study, appropriate radiation dose reduction strategies are implemented that optimize radiation doses for pediatric patients.

Different percentages (25%, 50%, and 75%) of the patients were administered particular doses of CTDIvol, and DLPs were calculated from the data.

Table 2. CTDIvol and DLP values for pediatric head CT scans in Abu Dhabi, UAE, with different percentiles; “y” represents year, and “N” represents number of patients.

		CTDIvol (mGy)				DLP (mGy· cm)
Age Group	N	25th	Median	75th	Range	25th	Median	75th	Range
<1 y	22.00	9.700	18.20	18.40	6.790–19.30	145.0	2700	304.0	92.20–3410
≥1 y <5 y	53.00	17.90	19.80	21.10	8.270–47.00	280.0	348.0	385.0	120.0–876.0
≥5 y <10 y	80.00	15.00	20.00	23.40	12.50–62.70	305.0	388.0	441.0	177.0–1300
≥10 y <16 y	91.00	14.90	23.40	27.80	11.20–57.20	386.0	471.0	568.0	389.0–1119

shows these results. As can be seen, 75% of the children below 1 year received 18.4 mGy of CTDIvol and 304 mGy·cm of DLP. These values increased, as expected, with older age groups. For example, for those aged 10 to 16 years, these doses reached 27.8 mGy and 568 mGy·cm. The increase in values for the CTDIvol and DLP with age in pediatric CT scans is likely due to the larger size of the older children and the need for higher radiation doses to penetrate more tissue. However, it is important to note that while larger pediatric patients may require higher radiation doses, efforts should still be directed toward minimizing radiation exposure as much as possible.

An important observation is a large variation in the DLP dose quantities for the head CT scan for the age group < 1 year. The large variation in DLP doses for head CT scans in infants younger than one year of age can be attributed to several factors such as patient size and shape, scanner settings, imaging protocol, and patient motion. Infants have smaller heads and necks, making it challenging to position them correctly, and are more likely to move during the scan. The imaging protocol and scanner settings can be adjusted by the technologist, which can lead to differences in dosage. These factors highlight the importance of careful planning and optimization of CT protocols to minimize radiation doses for this vulnerable patient population. For the other age groups, the DLP values ranged between 92.3 mGy·cm and 341 mGy·cm. In the group of ≥1 y–< 5 y, the range was between 120 mGy·cm and 876 mGy·cm. The age group of ≥5 y–< 16 y had a range between 177 mGy· cm and 1300 mGy·cm, and for the age group of ≥10 y–< 16 y, the range was between 389 mGy·cm and 1119 mGy·cm. A similar variation was observed in the CTDIvol dose quantities for head CT scans for all the age groups. Variations in pediatric head CT imaging protocols, e.g., whether the protocols are based on one sequence or more, image quality, or clinical indications (whether they are considered or not), can impact radiation doses, so even the same scanner could generate higher or lower values than DRLs.

Using the calculated data, the values at 75% were considered local DRLs (LDRLs). Figure 1 shows these results for both CTDIvol and DLP. The LDRL values of CTDIvol were between ~18 mGy and 27.8 mGy. This is within the recommended doses, as provided by Miglioretti et al. [16]. In this study, the authors analyzed data from a large healthcare system to establish DRLs for pediatric head CT scans. They found that the recommended DRLs for pediatric head CT scans should be no more than 50–60 mGy, depending on the age and size of the child. For example, the recommended DRL for a 1-year-old child is 50 mGy, while the recommended DRL for a 10-year-old child is 60 mGy. According to the results, the head CT scan doses given to the children in the relevant age groups in Abu Dhabi were all below the upper-limit values.

To further examine the local diagnostic reference limits in the Emirate of Abu Dhabi, we compared our results with the available data from other countries. The lack of international uniformity in age stratification for DRLs is a challenge that is often encountered in studies comparing DRLs in pediatric imaging between different countries. In addition, differences in patient characteristics, imaging protocols, and equipment performance may also affect the comparability of DRLs across different regions and countries.

Despite these challenges, the comparison of DRLs in pediatric imaging is important for identifying potential areas for improvement and optimization of imaging protocols.

We included only reference studies that considered similar age groups of children, ensuring a valid comparison among the data from different studies. Figure 2 shows these comparisons. As can be seen, the received doses were lower than those found in other selected countries. Abu Dhabi’s LDRL values are between 5% and 63% lower than those used in other countries, except for Turkey, where the values are lower than those of the UAE, by 5%. These statistics demonstrate that SEHA facilities follow the best protocols for pediatric imaging in accordance with the highest radiation protection standards.

Our study has a limited comparative component due to the small number of publications on pediatric CT scans and the lack of international uniformity for age stratification for DRLs. Due to this limitation, we were not able to make a concrete comparison between our data with the European guidelines for pediatric imaging [22]. Nevertheless, one can still present a rough comparison, as shown in

Table 3. Pediatric head CT local DRLs in Abu Dhabi in comparison with European DRLs; “y” represents year, and “m” represents months.

European Guidelines’ DRL			Abu Dhabi LDRL
Age Group	CTDIvol (mGy)	DLP (mGy·cm)	Age Group	CTDIvol (mGy)	DLP (mGy·cm)
0 < 3 m	24	300	<1 y	18.4	304
3 m < 1 y	28	385	≥1 y < 5 y	21.1	385
1 < 6 y	40	505	≥5 y < 10 y	23.4	441
≥6 y	50	650	≥10 y < 16 y	27.8	568

(despite the difference in age groups). The results indicate that Abu Dhabi’s local DRLs are comparable to or even lower than European DRLs. For example, for the youngest age group (<1 year), the local DRL in Abu Dhabi for CTDIvol is 18.4 mGy, while the European guidelines’ DRL in the two age groups of <3 months and <1 year are 24 mGy and 28 mGy, respectively. This suggests that the local DRL in Abu Dhabi is lower than the European guidelines for the youngest age group. In the age group of > 6 years, the European guidelines’ DRL is 650 mGy· cm, while in Abu Dhabi, in the two age groups of ≥5y < 10 y and ≥10 y < 16 y, the local DRLs are 441 mGy cm and 568 mGy·cm, respectively.

The comparison of DRL values among different countries is important for identifying the potential areas for improvement and for ensuring that the radiation doses delivered to patients are in line with international standards. While this study reveals that the LDRL values for CT brain procedures in Abu Dhabi are lower than those found in other selected countries, it is important to note that there may be variations in DRL values between different countries due to differences in population characteristics, imaging practices, and regulatory requirements. Although diagnostic reference levels (DRLs) provide guidance on the acceptable levels of radiation doses for specific imaging procedures, they do not provide a complete picture of the quality of pediatric CT imaging protocols. Other factors such as the appropriateness of imaging referrals, the use of dose-reduction techniques, and the quality of image interpretation can all impact the overall effectiveness of pediatric CT imaging. This requires collaboration between clinicians, radiologists, and technologists to ensure that pediatric CT imaging is performed in a safe, effective, and appropriate manner.

Overall, the results of this study are promising and provide significant reassurance to healthcare providers and patients in the region. The ongoing monitoring and refinement of imaging protocols will help to ensure that radiation doses remain as low as possible while still providing the necessary diagnostic information.

Finally, although the total number of patients considered in this study is not high, the number of patients in each age group is still larger than 10, which is the minimum number required to draw any concrete conclusion [12]. Thus, the results are statistically significant.

4. Conclusions

We surveyed the radiation doses administered to pediatric patients in terms of the CT dose index volume (CTDIvol) and dose–length product (DLP). The study’s primary goal was to collect CT dose index data from the largest SEHA CT scanner facilities and use them in the establishment of national DRLs. This survey provides a current snapshot of CT equipment technology and CT imaging practices in the Emirate of Abu Dhabi, UAE. Our results indicate that the doses received by different age groups in this region lie well within the guidelines and safety protocol used internationally.

The information provided indicates that the SEHA health system in the region offers low doses to pediatric patients, in the majority of cases. This study serves as a local reference and is a starting point for setting radiation dose protocols and national reference levels in the Emirate of Abu Dhabi and in the United Arab Emirates. For the purposes of establishing DRLs within hospital networks, the study results will be shared with the SEHA, the Ministry of Health, the Dubai Health Authority, and the Federal Authority for Nuclear Regulation (FANR).

The awareness of dose variations and associated risks is crucial for reducing radiation exposure. Using consistent imaging protocols and adjusting CT settings according to clinical indications may reduce variations in radiation exposure. Hence, a further investigation of scanning protocol variability, protocol optimization, and assessment of image quality in SEHA hospitals is planned as part of future research plans.

A regular review of DRLs and imaging protocols is particularly important in pediatric imaging, where children are more vulnerable to the harmful effects of radiation exposure. Adherence to established DRLs and safe imaging protocols can help to minimize the risks associated with pediatric CT imaging.

It is important to point out that this study also has several limitations. The data used were obtained from a single hospital in Abu Dhabi. This may have implications for the generalizability and representativeness of the study’s findings. However, as the SSMC hospital is the largest and leading healthcare provider in the region, we assume that the results are significant in providing insightful information regarding the radiation dose levels associated with pediatric head CT scans.

In summary, it is important to note that while adherence to established DRLs and safe imaging protocols can help to minimize the risks associated with pediatric CT imaging, there are still potential long-term effects that need to be considered. Therefore, it is crucial that healthcare providers take a cautious approach when ordering CT scans for children and consider alternative imaging methods whenever possible. By doing so, we can ensure that children receive the highest quality of care while minimizing their exposure to potentially harmful radiation.