Abstract
BACKGROUND:
To reduce radiation dose and subsequent risks, several legislative documents in different countries describe the need for Diagnostic Reference Levels (DRLs). Spinal radiography is a common and high-dose examination. Therefore, the aim of this work was to establish the DRL for Computed Tomography (CT) examinations of the spine in healthcare institutions across Jordan.
METHODS:
Data was retrieved from the picture archiving and communications system (PACS), which included the CT Dose Index (CTDI (vol) ) and Dose Length Product (DLP). The median radiation dose values of the dosimetric indices were calculated for each site. DRL values were defined as the 75th percentile distribution of the median CTDI (vol) and DLP values.
RESULTS:
Data was collected from 659 CT examinations (316 cervical spine and 343 lumbar-sacral spine). Of the participants, 68% were males, and the patients’ mean weight was 69.7 kg (minimum = 60; maximum = 80, SD = 8.9). The 75th percentile for the DLP of cervical and LS-spine CT scans in Jordan were 565.2 and 967.7 mGy.cm, respectively.
CONCLUSIONS:
This research demonstrates a wide range of variability in CTDI (vol) and DLP values for spinal CT examinations; these variations were associated with the acquisition protocol and highlight the need to optimize radiation dose in spinal CT examinations.
Introduction
Spinal pain is the most common reason patients attend physicians in the USA, with healthcare expenditures for spine problems totalling approximately $86 billion annually [1]. Many patients receive medical imaging, including X-rays, Magnetic Resonance Imaging (MRI), or Computed tomography (CT). CT examinations represent 50% of all man-made radiation delivered to humans [2, 3], with radiation doses per examination and frequency of examinations increasing rapidly. Increases are due to technological advances in technology and increased complexity of examinations [4]. However, cancer risks have long been identified with CT examinations, with reports suggesting up to 27,000 neoplasms being induced by CT exposure in a year [5]. There is limited evidence in Jordan or elsewhere that radiation dose reduction coupled with optimal image quality can be achieved.
Physicians may request a CT scan of the lumbar spine before or after x-rays. With submillimetre information obtained by CT scanners and the increase in the number of CT scans requested, the risk of inducing cancer with a CT examination of the entire lumbar spine is reported to be approximately 1 in 3,200, and in the thoracic spine (approximately 1 in 1,800). The increased risk with the thoracic spine is due to the additional coverage required and the anatomic implications of the cervical-dorsal junction CT scan range [6]. To reduce the potential risks associated with high radiation doses from CT, it is recommended that the “As Low As Reasonably Achievable”(ALARA) principle established by the International Commission on Radiological Protection (ICRP) [7] be applied.
DRLs are radiation dose values set, that should not be exceeded for average-sized patients when good practice is applied. They are usually protocol-specific for examinations and provide a reference level for that examination. DRLs are not dose limits but rather guidelines. However, if regularly exceeded, corrective action should be sought. DRLs were first introduced in the United Kingdom two decades ago [8]. They have been cited as effective tools for reducing radiation doses, with reports indicating a reduction of up to 50% since their adoption [9]. Also, in the absence of DRLs, dose variations for the same examination up to 23 times have been reported between centres for non-CT X-ray examinations [10] and CT examinations [11]. The dominant method of establishing a DRL is to determine radiation dose levels delivered for specific studies in several tertiary institutions in an individual country (or state) and then use these data to determine the 75th percentile of the dose distribution. Due to the varying procedures and equipment across different countries, most countries prefer to produce their own DRL rather than adopt levels established in other jurisdictions. The requirement for DRLs has been described in legislative documents in Jordan and throughout Europe, and their implementation has been observed in Europe and the United States [8, 12–16]. Still, there is no national DRL value for spine CT exams in Jordan. This research project aims to establish DRL for spinal CT and, in doing so, investigate the causes of variation and identify the lowest radiation dose required to facilitate the accurate detection of spine pathology by multi-detector CT.
Methods
Ethical approval
Institutional Review Board (IRB) approval (43/140/2021) was obtained, and the ethics committee exempted informed consent. Approval from the study institution’s medical director and IRB office was obtained before collecting data.
Study design and population
In line with a methodology previously described (NRPB, 1992), the DRL part of the study consisted of two primary data collections: 1) protocol data and 2) actual scan sequence data. Both parts involved sending out questionnaires to 30 centres across Jordan. Nine centres agreed to participate in the current work, and a clinical coordinator with CT experience was assigned to each centre to receive and administer the questionnaires. The data questionnaires and forms have been previously designed and validated [8]. In the first part of the questionnaire, information was provided on standard protocols used for specific examinations, including exposure factors, procedures, and radiation doses. The second part recorded the same data, but this time for protocols employed for each of the last 30 average-sized patients (60–80) kg [8] (Table 1). The number of centres chosen (n = 9) represents approximately 30 % of CT units in Jordan, exceeding sample sizes previously used to establish DRLs [8–10]. The study excluded patients younger than 18 years old. Documents were reviewed for each institution to determine if they had a quality assurance program in place.
For each scanning sequence, parameters were collected from each clinical centre
For each scanning sequence, parameters were collected from each clinical centre
CT scanners provide dosimetric indices for each sequence and examination, including Computed Tomography Dose Index (CTDI (vol) ) and Dose Length Product (DLP).
Examinations chosen
Currently, the research will focus on the most common adult spine investigations, cervical spine (C-spine) (n = 316) and lumbar-sacral spine (LS-spine) (n = 343). Thoracic spine studies were excluded due to a lack of sufficient cases.
Data analysis
The first, second, and third quartiles for CTDI (vol) and DLP were calculated using SPSS 22.0 (IBM Corp., Armonk, NY, USA). Each institution’s median radiation dose values were calculated, and the 75th percentile distribution of the median value of CTDI (vol) and DLP was used to define DRL. In addition to the above calculations, a stepwise regression was performed to identify exposure or procedural parameters associated with higher doses and to quantify each factor’s contribution to observed variations. An anonymised histogram, with the 75th percentile marked, will be circulated to each cipating centre. The provided information will enable each centre to identify the histogram bar(s) corresponding to their data, along with the specification of factors potentially responsible for higher doses. Additionally, international DRL levels were compared with the current national DRL levels. No other identifying features were presented.
Results
The patients demographic age range was from 18 to 92 years. In line with a methodology previously published, 659 protocol and scan sequence details were collected (Table 2).
Detailed information about each hospital’s scanners
Detailed information about each hospital’s scanners
The survey focused on the cervical and lumbar spine CT examinations, and CT scans included one single scan per examination, and repeated scans were excluded. To prevent variability in radiation dose recording, patients who weighed between 60 and 80 kg were included in the study [8].
Nine hospitals (four militaries, two private, and three governmental) participated in the study. Recordings from 659 (316 c-spine and 343 LS-spine s) CT examinations were collected. Of the participants, 68% were males and 31.6% were females, and the mean weight for both genders was 69.8 kg (min = 60; max = 80, SD = 8.9). Data was collected from nine scanners, of which three and five centres acquired image data in 64 and 16 slices. Only one centre used 128 slices. A sample size calculation indicated that a difference in dose of 5% at 0.8 power would be detectable. Table 3 shows patient and equipment characteristics as well as spine CT protocols.
Patients and equipment’s characteristics
IQR: Interquartile range.
The median CTDI (vol) and DLP per CT examination were collected for each site and used to compare doses across CT centres. Hospitals surveyed found wide variations in median CTDI (vol) between c-spine and l-spine examinations, with differences of 4.2 and 5.6 folds, respectively. DLP reported a similar pattern for the examinations surveyed, with differences of 3.2 and 5.4 for c-spine and LS-spine scans retrospectively (Table 4). Figure 1 illustrates the median CTDI (vol) and DLP distribution for the nine surveyed hospitals, with the horizontal line representing the DRL for both the cervical and LS-spine. The median CTDI (vol) values for the c-spine and LS-spine s were 5.63–23.8 and 6.43 –36.1 mGy, while the median DLP values for the c-spine and LS-spine were in the range of 184.6–679.9 and 266.9 and 1025.0 mGy.cm. Except in cases where some dosimetric values were doubled, there was no statistical significance between the centres. For example, the median CTDI (vol) value for the LS-spine for the hospital (5) was 6.4 milli-gray (mGy) compared with 36.1 mGy for the hospital (7). For the same centre, similar radiation dose values were observed for DLP values. Two hospitals have higher local median CDTI values (vol) than DRL; a third hospital has a slightly higher local median DLP value than DRL.
DRLs for the C-spine and LS-spine are stratified by centre
(A) Distributions of median CTDI (vol) values for c-spine CT examination. (B) Distributions of median DLP values for c-spine CT examinations. (c) Distributions of median CTDI (vol) values for LS-spine CT examinations. (D) Distributions of median DLP values for LS-spine CT examinations.
A summary of DRLs per scan (C-spine and LS-spine CT scanning examinations) and per centre is shown in Table 5. The DRL for the CTDI (vol) for the C-spine was 19.2 mGy and 565.2 mGy.cm for the DLP. Similarly, the CTDI (vol) for the LS-spine was 22.2 mGy and 976.8 mGy.cm for the DRL. Table 6 compares the findings with DRLs published in other countries [8, 12–16].
DRLs for C-spine and LS-spine CT scans in Jordan
C-spine and LS-spine CT DRLs in Jordan compared with other international: C-spine and LS-spine CT DRLs
Multiple regression analysis suggested that combining of kVp, mAs, and several slices accurately predicted CTDI (vol) more than any individual variable alone. The multiple linear regression results showed that mAs, kVp, and the number of slices were significant predictors of DLP. Table 7 shows that all factors were significant positive predictors.
The prediction of CTDI (vol) and DLP by using stepwise regression factors
We conducted a nationwide survey of delivered radiation for adult cervical and lumbosacral spine CT examinations to establish DRL in Jordan. For the first time, our survey revealed information on the radiation doses of adult patients from spinal CT (median and 75th percentile for the DLP of each anatomical area). These values represent the current CT radiation dose levels and activities in Jordan. We collected data about the radiation doses according to different parameters, the type of institution, the type of CT scan manufacturer, and the number of CT detectors. Then, we benchmarked our doses with internationally reported DRLs. When compared with six international countries, Jordan demonstrated high CT radiation levels. These doses should raise concern and attention from health professional organisations to investigate their causes and establish methods for optimisation to achieve radiation reduction. This research demonstrates a wide range of radiation dose variations for spinal CT scans across the hospitals. With a range of 184.6–679 mGy cm, the highest median DLP was 3.7 times higher than the median value. Multiple regression analysis predicted that low DLP depended on mAs, kVp, and the number of slices.
We recommend that radiology centres, government agencies, and radiographers develop patient dose management systems and quality improvement programs. Radiology departments should provide radiographers with continuous training to enhance their knowledge of radiation protection, protocol justification, and optimization methods to answer clinical questions with minimal radiation dose while maintaining an acceptable level of image quality for accurate radiological diagnosis. Medical imaging personnel must also audit and monitor their radiation dose levels regularly, comparing them with Jordan’s established DRL. A new CT technology and iterative image reconstruction algorithms are also valuable. All radiation emitting units in government institutions should be subject to mandatory audits [17, 18]. Furthermore, they should follow up on regional and DRL in Jordan annually, considering all medical imaging procedures. Moreover, physicians should always consider the risk-benefit ratio when referring patients for imaging, which can only occur through radiologists’ intervention, and radiographers should follow ALARA principles, study justification, protocol optimisation, and radiation protection.
Ultimately, size-specific dose estimates (SSDE), including patient diameter measurements to correct CTDI (vol) for patient size, are the most accurate form of radiation dose received and how it affects the organs specifically. This was introduced by the AAPM and implemented in the United States of America national DRL [19–23]; however, SSDE is not part of the current Jordanian DRL procedures. Further work is recommended to address achievement for radiation optimisation. There are shortcomings in our study. Firstly, effective doses were not calculated, which are good measurements to establish DRL. Secondly, we didn’t record different established protocols for each anatomical area, which may vary from institution to institution. Thirdly, no control for the patient’s height was established, which may influence the DLP values. Fourthly, we did not consider the different reconstruction algorithms employed in other institutions that potentially affect radiation levels. Finally, we did not assess image quality, which was beyond the scope of our survey. However, further work should investigate image quality to achieve optimisation of CT examination.
In conclusion, the radiation doses resulting from CT examinations result in radiation doses with proven risks of inducing cancers. This work addressed the requirement to reduce CT doses stated by legislative, regulatory, scientific, and professional bodies. It will benefit the millions of individuals who undergo this procedure in Jordan and elsewhere each year. This study marks the establishment of the first National Diagnostic Reference Level (NDRL) for adult spine CT scans in Jordan. The established Jordanian DRLs for CT scans of cervical and lumber were higher than most international doses. The differences were associated with variations in the mAs, kVp, and slice number. This demonstrates the importance of DRLs in Jordan to assist in the optimization of CT scanning procedures and protocols.