Authors: Nickelas Huffman (*College of Medicine and Life Sciences, The University of Toledo), Ajay Nair (*College of Medicine and Life Sciences, The University of Toledo), Summer M. Drees (*College of Medicine and Life Sciences, The University of Toledo), Trevor Bouck (†Department of Orthopaedic Surgery, The University of Toledo Medical Center), David Yatsonsky (†Department of Orthopaedic Surgery, The University of Toledo Medical Center), Kerry Krugh (‡Department of Medical Physics), Sara Seegert (§Research Department), Benjamin H. Russell (*College of Medicine and Life Sciences, The University of Toledo), Adrian Lewis (∥Department of Pediatric Orthopaedic Surgery, Russell J. Ebeid Children’s Hospital, Toledo, OH), Aaron Buerk (∥Department of Pediatric Orthopaedic Surgery, Russell J. Ebeid Children’s Hospital, Toledo, OH), Gregory M. Georgiadis (¶ProMedica Physicians Orthopaedic Trauma and Adult Reconstruction, ProMedica Toledo Hospital)
Categories: Trauma, pediatric radiation exposure, computed tomography, effective dose, ionizing radiation
Source: Journal of Pediatric Orthopedics
Authors: Nickelas Huffman, Ajay Nair, Summer M. Drees, Trevor Bouck, David Yatsonsky, Kerry Krugh, Sara Seegert, Benjamin H. Russell, Adrian Lewis, Aaron Buerk, Gregory M. Georgiadis
Children who sustain femur fractures are exposed to medical radiation as part of their treatment. In addition to standard radiographs and fluoroscopy, computerized tomography (CT) is a major source of ionizing radiation (IR). These patients frequently undergo CT scans during their treatment. The aim of the current study was to quantify the factors associated with greater IR exposure in children with operatively treated femur fractures and compare radiation exposure among those who received CT scans versus those who did not.
Thirty-eight patients below 18 years old with operatively treated traumatic femur fractures were included in the study. They fell into one of 2 1 received 1 or more CT scans (CT group), and the other did not receive a CT scan (non-CT scan). The 2 groups were then compared by total quantitative radiation exposure, age, BMI, location of the femur fracture, open versus closed fracture, and the presence versus absence of fracture displacement.
The effective dose of IR exposure that occurred within the operating room (OR) made up 75.7%±27.4% of the total IR exposure among all patients. Patients in the CT group had over a 5-fold greater quantitative cumulative IR exposure compared with the non-CT group (P<0.0001). Furthermore, patients in the CT group were significantly older than (P=0.004) and had a greater BMI (P=0.045) than the non-CT group.
Children with pediatric femur fractures often exceed what is considered a low level of radiation. There is a significant difference in radiation exposure between pediatric femur fracture patients who receive 1 or more CT scans compared with those who do not receive a CT scan.
Level III—retrospective cohort study.
Orthopaedic trauma relies heavily upon imaging techniques. These include plain radiographs and advanced imagery. Once fractures have been identified, serial radiographs may be needed.^1^ Polytraumatized children often receive computed tomography (CT) scans which are a major source of ionizing radiation (IR).^2,3^ In addition, intraoperative fluoroscopy is used for surgical fracture management. Children are more sensitive to ionizing radiation than adults.^4^ Pediatric patients’ long postexposure life expectancy also contributes to the increased risk.^1,5,6^
The total pediatric radiation exposure for the entire episode of care for pediatric femoral shaft fracture has not been studied. We chose this particular fracture because it often requires surgery, multiple x-ray exposures, and is often associated with polytrauma. Previous investigators have suggested that patients with proximal femur fractures were exposed to the highest amount of IR.^7^ This major long bone injury often leads to numerous preoperative and postoperative radiographs, fluoroscopy for implant insertion and subsequent removal, and additional studies like head or abdominal CTs or skeletal surveys.
IR exposure is a known health hazard.^2,6,8^ Potential sequelae include the development of dermatitis, erythema, leukemia, brain, breast, and other cancers.^5^ Understanding the extent of radiation exposure can raise awareness and ultimately lead to approaches to minimize it. To the best of our knowledge, this is the first study that quantitatively examines radiation exposure in pediatric femur fracture patients for their entire episode of care. We compared the total radiation exposure of pediatric femur fracture patients who undergo 1 or more CT scans with those who do not undergo CT. Our goal was to quantify factors associated with greater IR exposure in this subset of injured children.
This was a retrospective study examining the total amount of radiation exposure for children who presented to our Level II pediatric trauma center with a femur fracture requiring operative repair between January 1, 2017 and December 31, 2020. Total radiation exposure included fluoroscopy for hardware insertion and removal, all diagnostic preoperative and postoperative imaging, in addition to all follow-up radiographs. The study was approved by the Institutional Review Board and a waiver of consent was obtained.
Inclusion criteria consisted of age below 18 years old, nonpathologic femoral shaft fracture treated with internal fixation, and follow-up within our institution. This included initial surgery, all follow-up visits, and implant removal when performed. Patient variables collected included age at injury, sex, BMI, as well as the presence of an open fracture.
Patients were excluded if full treatment records were inaccessible or postoperative care was completed outside our healthcare system. Also excluded were fractures from nonaccidental trauma.
The hospital quality department and the trauma quality databases were queried for the ICD-10-CM code S72.3 (fracture of the femoral shaft). Information obtained was cross-referenced and additional data was abstracted from the electronic medical record. Patients were excluded if the fracture occurred above the lesser trochanter or was contained within 5 cm of the distal femoral physis/physeal scar. Records were reviewed for all IR related to the injury through completion of treatment, with a minimum of 1-year follow-up. Participants were separated into 2 a group that received at least 1 CT scan (CT group), and a group that did not receive a CT scan (non-CT group).
Statistical analyses were conducted utilizing the IBM SPSS v25 software. Continuous variables were compared using the Student t test while categorical variables utilized the χ^2^ test. Results are reported as mean±SD, and statistical significance was determined as having a P-value <0.05. A PhD hospital radiation physicist performed the dose calculations, and a statistician analyzed the data.
Effective dose (E) for each individual CT scan was extracted from commercially available automated dose monitoring software, Radimetrics Enterprise Platform, version 3.0b (Bayer HealthCare). This database software intakes Radiation Dose Structured Reports (RDSR) directly from the CT scanners following each exam. The software then parses information related to the patient, scan parameters, and radiation dose from the RDSR and uses this information to calculate E using ICRP 103 weighting factors. The E calculation accuracy with this software has been confirmed in the literature.^9^
The E was calculated for each radiographic view using the methods of Huda and Gkanatsios.^10–12^ This is a multistep process
The ε calculation^10^ takes into account various radiographic factors, notably the peak kilovoltage (kVp), tube current (mAs), and field size. Considering these radiographic factors were adjusted for patient size, each individual patient was assigned to 1 of 5 size categories based upon the patient's BMI. For a given radiographic view, the typical radiographic factors were determined for each size category and the subsequent ε was calculated.
Conversion factors (E/ε) that have been established in the literature^11,12^ were used to convert ε to E for a standard adult patient.
The ED for a standard adult patient was scaled to an ED for each pediatric patient using a factor based upon the patient mass. This methodology has been established in the literature.^11,12^
Sixty-three pediatric femur fractures were found in our trauma registry. Thirty-eight (31 males and 7 females) met the inclusion criteria. Ages ranged from 3 to 17 years (10.6±4.3). Most femoral fractures under 5 years of age are treated without internal fixation, but one 3-year-old in this series with a widely displaced fracture underwent elastic nailing. The cohort included 24 Caucasian individuals, 12 Black individuals, and 2 individuals of another race. Mechanism of injury (MOI) included motor vehicle accident (MVA) (n=15), sport-related (n=12), fall (n=6), and other (n=5). All subjects underwent surgical repair with internal fixation. All but 1 patient had a closed fracture. Thirty-three patients experienced femur fracture of the mid to distal shaft, and 5 patients experienced femur fracture of the proximal shaft.
All fractures healed and the minimum follow-up was 1 year. Twenty-two of the 38 patients had their implants removed. There was incomplete data on fluoroscopic radiation exposure at the time of hardware removal in 12 of the 22 patients. Of 14 patients with complete records, all but 1 had less exposure during implant insertion than removal. One patient with a locked intramedullary nail received more fluoroscopy during plant removal than during insertion. Fluoroscopic exposure was recorded in milligrays (mGy). The average was 8.75±8.27 mGy (range: 1.04 to 33.49) on implant insertion and 5.82±8.91 mGy (range: 0.04 to 27.45) on removal.
IR exposure was calculated for all plain radiographs, CT scans, and available surgical fluoroscopy date in the electronic medical record. Mean IR exposure was 1.3±9.0 mSv for all patients from plain radiographs. On average, x-rays made up 7.2%±16.6% of the total effective dose of IR among patients in the cohort. Fifteen patients underwent CT scans, from which the mean total IR exposure was 28.8 mSv with an SD of 15.8 mSv. The surgical effective dose of IR exposure mean was 26.0±37.1 mSv. On average, fluoroscopy obtained within the OR made up 75.7%±27.4% of the total effective dose of IR exposure among all patients. The total mean exposure for all subjects from all sources of radiation during the total time period of the study was 38.0±49.24 mSv.
There were 15 patients who underwent at least 1 CT scan, and 23 patients who did not receive a CT scan. CT scan in the group with injuries in addition to the femur fracture included CT scan of the brain, spine, chest, face, and CT arterial portography (CTAP) for the evaluation of head trauma or blunt abdominal trauma.
The number of CT studies was counted based on parts of the body scanned. Thus, if a pediatric trauma patient was scanned from head to pelvis, this could have counted for up to 4 studies (brain, cervical spine, chest, and abdomen). Only one of the 15 patients scanned underwent a study of a single anatomic location. The remainder had multiple body parts investigated (3 patients had 5 studies, 10 had 4 and 1 patient had 3).
CT scans were ordered by the emergency room or the on-call trauma attending. All were performed during the initial admission. Most occurred in the initial emergency room trauma assessment, but some were performed later in the child’s admission to reassess for brain, facial, or bowel trauma.
When comparing the 2 groups (CT in isolated femur fracture vs. CT in patients with associated injuries), they did not differ in sex (P=0.06), fracture location (P=0.70), or open versus closed fracture (P=0.39) (Table 1).
There was a significant difference between the 2 groups in total radiation exposure (P<0.0001) (Table 1). There were also significant differences in age and BMI. Eight patients had injuries in addition to their femur fracture. These included brain, chest, spine injuries, and other long bone fractures. All had CT scans. Seven isolated femur fracture patients also underwent CT scans.
The 5 proximal femur patients averaged 39.6 mSv of radiation exposure, which was lower than the average of patients receiving CT scans (75.2), but higher than the non-CT group (13.7). Overall, 15 of the 38 patients in our study exceeded 20 mSv of radiation exposure.
Physicians should be aware of the cumulative effect of radiation exposure over time.^7^ The aim of the current study was to provide a descriptive analysis of the total radiation exposure in pediatric femur fracture patients. This injury was selected because of its generally operative nature, necessitating significant initial fluoroscopic exposure, serial radiographs, and repeat fluoroscopy during potential implant removal.
In the current study, a substantial portion of IR occurred during fluoroscopic use in the operating theater. This is consistent with previous studies. Decreasing fluoroscopy time during pediatric intramedullary nailing has been recommended.^13^ Significant variability in IR has been noted during fluoroscopy of pediatric supracondylar distal humeral fractures. While some factors were not found to affect radiation exposure, other differences such as fracture severity, pin number, and the operating orthopaedic surgeon did influence the amount of radiation exposure.^14,15^
The current study documented the amount of additional IR that occurred with adjunct CT scans.
The amount of radiation incurred from a single scan can be quite variable depending on the number of body parts scanned. In our study, the mean exposure in the CT group was 28.8 mSv with an SD of 15.8. These patients had greater than a 5-fold increase in the average cumulative effective dose of radiation exposure. For the pediatric population, <20 mSv is considered a low level of IR.^16^ Many of the patients (18 out of 38) exceeded this threshold. One study found that pediatric patients undergoing diagnosis and treatment of long-bone fractures remain within this low level of IR.^6^ Another found evidence that 2 to 3 CT scans in pediatric patients resulted in an effective dose of IR that ranged from 30 to 90 mSv,^17^ and exposure to an effective dose of IR within this range has been found to be associated with an increased risk of cancer in pediatric patients.^16–21^
The cumulative radiation exposure in the non-CT group was in the range of 0.198 to 61.23 mSv. The exposure in the CT group ranged from 11.47 to 191.60 mSv, which was within the range of significantly elevated IR.^18^ The differences between the 2 groups reached statistical significance (P<0.0001).
Patients receiving CT scans were older (nearly 50% were above 15 years old) and had a BMI greater than the non-CT group. This may suggest that physicians were aware of the risks of radiation exposure and were less likely to order a CT scan in younger age groups. The risk of thyroid, breast, brain, leukemia, and nonmelanoma skin cancer were related to the effective dose of radiation, and these risks appeared to be the greatest in children exposed to IR earlier in life.^1^ The effective dose of radiation absorbed during whole body CT imaging was found to be less in patients aged 14 years old compared with patients aged 4 years old.^22^ Future studies should aim to evaluate physicians’ attitudes towards ordering repeat CT scans for pediatric fracture patients.
Adult patients with a higher BMI have been found to be at increased risk of lower-extremity injuries in frontal crashes.^23,24^ While the effect of obesity on bone mineral density is site specific,^25^ 1 pediatric study found that excessive body fat was negatively associated with bone mineral density.^26^ This raises the possibility that greater BMI is associated with more severe femoral fractures, necessitating CT to exclude other injuries. However, there is limited evidence regarding a greater need for CT scans in pediatric patients with greater BMI.
Early consultation with the most senior member of the team would seem advisable to help minimize IR. A study of pediatric patients who underwent percutaneous pinning of distal humerus fractures found an association between attending surgeons with <1 year of experience and higher fluoroscopy time and radiation.^27^ Another investigation found no association between radiation exposure and surgeon experience.^6^ The current study did not record surgeon experience.
Previous studies have also discussed the importance of determining specific indications for when a CT scan is appropriate.^5^ For example, in pediatric pelvic fractures it has been suggested that plain radiographs can classify and manage most fractures.^28,29^ Key concepts to decrease IR include justification and optimization.^8^ Radiation safety programs may also be beneficial in raising awareness and decreasing radiation levels.^30,31^
The FDA has begun an initiative entitled “Reduce unnecessary radiation exposure from medical imaging” that includes establishing “reasonable scan protocols.”^3^ Ideally, tools to track and report a patient’s lifetime IR exposure should be readily available.
This was a retrospective analysis of 38 patients from a single hospital. It has many limitations. The small numbers of patients with disproportionately large IR burden in the CT group could skew IR measurements. Total radiation exposure was underestimated in many patients as the fluoroscopy measurements at the time of implant removal were incomplete. These results may not be representative of other institutions because of distinctive imaging protocols, variable use of fluoroscopy, different implants, varying rates of implant removal, and different surgeons providing operative care.
Comparison among providers was not performed to assess if fluoroscopic IR varied significantly between surgeons (although CT-based IR overwhelmed all other sources). Finally, the definition of a child can be open to interpretation. Our study included patients under 18 years of age. Acceptable doses of radiation may be variable across this age spectrum (eg, exposure to younger patients would be expected to have more harmful effects than older ones).
In the future the amounts of ionizing emissions from CT scans can be expected to decrease with improvements in technology. Nevertheless, efforts should always be made to limit medical radiation in children.
Children with operatively treated femur fractures may be exposed to high levels of radiation.
This study adds to the growing body of literature that emphasizes the need to minimize radiation in pediatric trauma patients.