ANALYSIS OF MRI UTILIZATION IN PEDIATRIC PATIENTS IN SAUDI ARABIA

1. Department of Radiological Sciences, Faculty of Applied Medical Sciences, Tabuk University, Saudi Arabia. 2. Department of Radiology, King Fahad Specialist Hospital, Tabuk, Saudi Arabia. ...................................................................................................................... Manuscript Info Abstract ......................... ........................................................................ Manuscript History Received: 30 November 2019 Final Accepted: 31 December 2019 Published: January 2020

1030 Tomography (CT) scan. A CT scan can rapidly diagnose many serious conditions and can provide valuable information regarding internal organs and vessels. However, the main disadvantage of CT scan is the relatively high radiation dose that a patient receives, which may increase the number of malignancies related to diagnostic imaging [1][2][3]. Despite this, approximately four million pediatric CT scans of the head, chest, abdomen, pelvis and spine are performed around the world annually, and are predicted to cause 4,870 future cancersfor an estimated 4 million patients [3]. Radiation from CT scanningis estimated to cause 0.7% of total expected baseline cancer incidents and 1% of cancer mortality [2]. Physicians, therefore, need to decide whether the risk of developing cancer as a result of radiation exposure outweighs the risks of not performing a CT scan at all.
Magnetic Resonance Imaging (MRI) is the most favourable alternative as it utilizes non-ionizingform of electromagnetic radiation. MRI is usually used for conditions that cannot be accurately assessed by other cheaper, faster imaging modalities such as plain-film x-rays, ultrasound and CT scanning. However, MRI has a much longer scanning time, which may increase the need to use anaesthesia in medically unstable patients. In addition, it is contraindicative in patients who have some types of metallic implants and who suffer from claustrophobia.
The use of pediatric MRI in radiology departments is poorly understood. Understanding the usage patterns and workflow of pediatric MRI would be a research area of great interest. The purpose of this study is to evaluate MRI usage for pediatric patients aged 13 years and under. Furthermore, based on the types of scans being performed, a radiologist can determine whether a pediatric patient needs an MRI scan, or whether they can be assessed by an alternative imaging modality.

Setting:
Ethical approval was obtained from the Institutional Review Board (IRB) committee at the University of Tabuk to conduct this retrospective study at single-site public hospital. The confidentiality of patients was ensured by saving all the relevant data on a computer that only the researcher had access to. In addition, all the patients taking part in the study were anonymised.
The MRI images collected using Optima 450w wide bore 1.5T MRI scanner, manufactured by GE Healthcare. Two scanners operate between 8 am and 5 pm and are used by the outpatient, inpatient and emergency departments. Thursdays are assigned for patients who usually need anaesthesia to remain still -such as patients who were less than two years old. Priority is usually given to patients referred from the emergency department or for urgent inpatient cases. The scanners stop operating at 5 pm and there are no night shifts required.
Outpatient MRI exams performed during this period are ordered by the pediatric department or referred from other hospitals. All requests are discussed with the radiologist on duty.

Data extraction:
Data from all pediatric MRI exams performed between January 2018 and January 2019 were obtained from RIS and were combined with demographic data.The total number of all MRI exams for all ages in the same period was also extracted for comparison. The data collected included age, gender, admission type, scan type, indication, and the use of contrast in order to find the extentof pediatric MRI Utilization.
The KFSH began operating in 2016 and has 500 beds. In addition to two other public hospitals, it serves a population of 800,000 in the North West province and surrounding areas. This study seeks to address the usage of pediatric MRI and to highlight whether it is being over-or underused. The purpose of our study was to evaluate the Utilization of MRIfor pediatric patients aged 13 years and under which are the ages that referred from pediatric clinic.

Results:-
There were 158 pediatric MRI exams performed on 121 patients aged 13 years and under (mean= 4.5 ± 4.8.) during the 12-month period (an average of 1.3 studies per patient). Five exams were excluded due to incomplete data meaning that 153 exams were analysed. The total number of MRI scans performed at the hospital over the 12-month period was 3,089, meaning that pediatric MRI exams accounted for 5.1% of the overall scans. For comparison, during the same time period, 2931 MRI exams were performed on patients older than 13years old.

1031
There were a slightly higher number of MRI exams performed on males (51.0%, n= 78 exams) than females (49.0%, n= 75 exams).Of those patients, 75 (49%) were inpatients (see Table 1) and 77 (50.3%) were outpatients, with only one patient being referred from the emergency department (see Figure 1). 1032 Figure 1:-The distribution of male and female exams between outpatients and inpatients.
The results showed that there were higher rates of female exams among inpatient (47/75= 62.7 %) and higher rates of male exams among outpatients (49/77= 63.6 %). In addition, 69.3% of the MRI exams were neuroradiological exams -of which 93.4% were of the brain and 6.6% were of the spine -21.6% were musculoskeletal and 4.1% were of the chest, abdomen and/or pelvis. Additionally, 15.7% (n= 24) of the exams used contrast media (gadolinium)-of which 77.1% were brain, 22.9% were musculoskeletal (see Figure 2). 1033 Within the inpatient group, the distribution of the procedure type is shown in Figure 3, where more females than males had brain, spine, abdomen and pelvis scans, while more musculoskeletal exams wereperformed on males. However, within the outpatient group, more males than females were scanned for each procedure (Figure 4). 1034 For inpatient scans the female-to-male ratio was 1.6:1; for outpatient scans the male-to-female ratio was 1.7:1; for neuroradiology scans the female-to-male ratio was 1.8:1; for musculoskeletal scans the male-to-female ratio was 2:1; and for all chest, abdomen and/or pelvis scansthe male-to-female ratio was 1.3:1. In addition, for exams requiring the use of gadolinium, the ratio of male-to-female was 2.6:1.
The distribution of age categories is shown in Table 2, where the most populous group was those patients who were 2 years old orless (n= 74), which represented 48.4% of the total pediatric MRI patients. In contrast, the least populous group were those patients aged between 5 and 6, with only seven scans performed between them.Among those patients aged 2 years or less, 49 of the 74 patients were female and the other 25 were male (see Figure 5).  1035 Table 3 shows the clinical indications for the exams and their frequencies. The most common indications for neuroradiology exams were headache, which accounted for 24.5% of all neurological exams, followed by Hypoxic Ischaemic Encephalopathy (HIE) with 17%. The most common indication for a spinal exam was back pain, which accounted for 25.1% of the scans. The most common indication for musculoskeletal scans was generalised pain, which accounted for 21.9% of all the musculoskeletal scans, while undescended testes were the most common indication for chest, abdomen and/or pelvis scans with 62.5%. Our results show that the inpatients who had MRI scans here were significantly younger (mean age = 2.40 std = ± 4.1) than the outpatients (mean age = 6.6, std = ± 4.6), where p = 0.000. The mean age in each exam type is shown in Table 3, where it can be seen that the average age of patients having brain scans is much less than those having abdomen and pelvis, musculoskeletal and spinal exams.

Discussion:-
In this study, we provided a report of how pediatric MRI is utilised in the radiology department of a single public hospital duringa 12-month period. A total of 69.3% of the exams were neuroradiology exams, which included the brain and spine. In previous reports, 98% of MRIs in the adult emergency department were neuroradiology exams [4], while 93% of the pediatric emergency MRI exams were also neuroradiological exams [5]. The higher rates of neuroradiological cases may be due to the fact that the cases from theemergency department are more urgent compared to the cases in our study, which mostly concerned outpatients and inpatients. However, there were lacks of pediatric MRI Utilization studies that compare inpatient and outpatient exams.
Our study shows a slightly higher overall rate of male patients (51.0%, n= 78 exams) compared to female patients (49.0%, n= 75 exams) who had an MRI scan. However, among inpatients, there were higher female patients than male. On the other hand, among outpatients, there was a higher ratio of males to females.The association between gender and type of admission was not well understood, and in order to provide more clarification, we analysed the association between gender and exam procedure within each admission type. Surprisingly, in outpatient exams, ratios of male MRI patients surpassed those of females in all the scan categories (neuroradiology, musculoskeletal and abdomen and pelvis scans). Within inpatient exams, on the other hand, the ratios of female scans were higher in all exam procedures except for musculoskeletal MRI. This raises the question as to whether female diseases are more critical than those of males, which require more care and therefore justify being admitted.
With regard to scan type, the overall pediatric MRI Utilization shows that there were higher ratios of male-to-female patients in spinal (1.3:2), musculoskeletal (2:1) and abdomen and pelvis exams (1.3:1). However, female ratios were higher than males in brain exams (1.2:1).
The reasons for the dominance of male ratios in all procedures except brain exams is still unknown; however, in contrast to the previous study where the neuroradiology exams were the largest contributor of the gender differences [6], musculoskeletal exams were the largest contributor to gender differences in our study.
We also analysed the association between gender and ratio of different categories and found that the female-to-male ratio was 1.6:1 for inpatient scans. For outpatient scans, the male-to-female ratio was 1.7:1; for neuroradiology scans, the female-to-male ratio was 1.8:1; for musculoskeletal scans, the male-to-female ratio was 2:1; and for all chest, abdomen and/or pelvis exams the male-to-female ratio was 1.3:1. For exams requiring the use of gadolinium, the male-to-female ratio was 2.6:1.
In our study, gadolinium contrast agents were used in 15.2% (n= 24) of the total cases. Although Gadolinium Based Contrast Agents (GBCAs) are considered to be a safe contrast agent, the FDA has recommended more studies to investigate the implications of GBCAs by conducting studies on GBCA Utilization in pediatric patients and to evaluate the agents and their doses in order to assist in the design of guidelines for the use of contrast indicated population [7].This is due to the fact that GBCAs remain in the body in measurable amounts in vital tissues such as the brain, despite the normal renal functions [8][9][10]. In addition, some studies reported that contrast can remain in bone tissue [11,12] and can be still be detected several years after administration [11]. Although no long-term adverse effects have yet been proven, GBCAs can nevertheless remain in small amounts in adults and children for years. Therefore, prudent adviceisto use GBCAs with caution [13].
In our study, musculoskeletal exams made up20.9% of the total sample. Although they were few in number overall, most of the musculoskeletal MRI exams were associated with male patients with a male-to-female ratio of 2:1.The most common indications weresuspected osteomyelitis followed by pain and trauma, respectively. In contrast, previous studies reported that the most common musculoskeletal indications in pediatric male patients were related to sports injuries [14] acute osteomyelitis and septic arthritis [15] as well as a greater prevalence of bone tumours [16].  [17].
Our study showed that only one patient was referred from the emergency room for an MRI scan.This may be due to the fact that the MRI scanners do not operate after working hours and could also be due to restrictions in the protocols of the radiology department that discourage emergency doctors from requesting MRI exams unless absolutely necessary. In addition, the CT scanner operates 24 hours a day and has the added advantage of having a rapid scan time, which allows for sedation-free scans and offers a fast and accurate alternative to MRI. Indeed, there are several factors that mean that MRI is notthe ideal modalityfor diagnosing emergency patients. These include a long waiting list, a long scan time, the possible need for sedation and the lack of staff outside of working hours. This was consistent with aprevious study performed in Saudi Arabia, where only 1.2% of all MRI exams for both adults and children were referred from the emergency department. Moreover, in another study, just 0.96% of all pediatric emergency patients had an MRI scan [18] and 0.51% in a further study of mixed adult and pediatric visits to the emergency department had an MRI scan [16].
MRI usage generally varies with age, where the highest rates were associated with patients aged < 1 year (31.6%, n=50) and 13 ≤years (8.8%, n= 14). This may be due to the need to reduce the exposure to ionising radiation for patients aged less than one year and due to other factors, such as 13-year old patients being more likely to be more stable, less claustrophobic and more likely to comply with instructions,while patients aged 6 tended to have the lowest number of scans (1.3%, n=2). This is in contrast to the date found in other studies, where the lowest number of scans was reported for those patients aged less than 3 years [17].
Brain MRI scans were significantly associated with youngest patients,with a mean age of 3.27 ± 4.6, while the highest mean ages were found in patients having spinal MRIs (mean age = 7.4 ± 4.9, p = 0.020) and musculoskeletal MRIs (mean age = 7.1 ± 3.9, p = 0.018). This may be explained by the fact that children become more active as they grow older, meaning that they are more likely to sustain a traumatic injury.
There was a significant difference between ages of inpatients (mean age = 2.4 ± 4.1) and outpatients (mean age = 6.6 ± 4.6), where p = 0.000, and where younger patients were admitted to hospital in higher numbers than older children. This is may be due to the fact that physical injuries such as the musculoskeletal and spinal traumas increases as the age of children increases which decreases the need for admission among older children. In this case other modalities such as x-ray and CT may provide adequate diagnosis for those patients and decrease the need for MRI. This was also related to the distribution of different age groups between different types of exams,as 99/153 (64.7%) of the total cases were for brain scans, with a mean age of 3.27 ± 4.6.
Limitations:-1. The results of this study were limited to a single public hospital in Tabuk City, Saudi Arabia. Therefore, extending this study to several hospitals in different regions in Saudi Arabia is highly recommended in order to get a more generalized overview of the Saudi population. 2. The data were only collected over a12-months. In order to more accurately evaluate trends in pediatric MRI exams, several years' worth of data will need to be collected. 3. Although we reported clinical indications for each exam, we did not link these scans to other modalities, and we did not conduct follow-ups with any of the patients. 4. This study is limited by incomplete data on medical history and exams as well as the treatment outcomes. 5. The hospital protocols concerned with managing the MRI requests were not considered.

Conclusion:-
The vast majority of pediatric MRI investigations were neuroradiological and were performed on younger children. There was a significant difference in the mean ages of inpatients and outpatients in addition to the variations in the mean ages among different exam types. This study shows that the age of the patient is the factor that most significantly affects the Utilization of MRI in children.