Discussion
In this rapid review, designed to synthetize the published evidence on potentially low value practices in orthopeadic trauma, we found consistent evidence of lack of effectiveness or harm for the following practices: routine initial imaging of the spine (<65 years of age), pelvis, knee and ankle in patients negative on a validated clinical decision rules or physical exam; preoperative blood tests in ASA grade I patients; spine consultation for isolated thoracolumbar transverse process fracture; immobilization with an orthosis in patients <60 years of age with A0-A3 thoracolumbar burst fractures; immobilization with a cast or splint for suspected scaphoid fracture negative on MRI or confirmed fifth metacarpal neck fracture; and routine follow-up X-ray in patients with clavicle, hip, femur, tibia, ankle or metatarsal fractures.
Studies on initial imaging of the cervical spine, pelvis, knee and ankle, consistently observed very small proportions of missed fractures in patients at low risk on validated decision rules. In addition, performing preoperative blood tests in ASA grade I patients did not lead to any adjustment in therapies. However, these results were largely based on systematic reviews of observational studies. Thus, high quality RCTs may be needed to confirm that these practices are low-value. Evidence for wrist imaging was less consistent with one study observing 10% of missed injuries.87 However, this evidence was based on three prospective studies conducted in patients with similar types of fractures but using different decision rules.40,54,87 This suggests that more rigorous research is needed to identify the most accurate clinical decision rule for initial imaging of the wrist.
Studies on initial consultation and therapeutic interventions provided consistent evidence of low-value care. Lack of spine surgeon consultation for isolated T1 to L5 transverse process fracture was not associated with neurological deficit. Importantly, none of the patients in these studies required either surgery or an orthosis up to 6 months post-injury, which are two treatments overseen by spine surgeons. Thus, these patients could potentially be discharged with recommendations for pain management and return to activities without a spinal surgery consultation in the presence of sufficient expertise in radiological interpretation. We found high-level evidence from systematic reviews of RCTs and RCTs suggesting that orthosis in patients <60 years of age with a A0-A3 thoracolumbar burst fracture and no neurological symptoms is associated either with no benefit or with harm. Orthosis is often prescribed to improve patient comfort but restricts patients’ movements causing discomfort and compromising sleep, which may explain why patients with no orthosis had less or equivalent disability. Similarly, we found high-level evidence from RCTs that immobilization of the fifth metacarpal neck fractures leads to greater or equivalent disability while immobilizing suspected scaphoid fractures is associated with greater disability. Thus, evidence suggests that limiting the mobility of patients with these injuries is not beneficial and may even cause harm.
Some inconsistencies in evidence were observed for post-treatment routine imaging and follow-up consultations. Fractures to the upper extremities, particularly fractures to the radius shaft and metacarpal were associated with high rates of immobilization prolongation in two studies in which routine imaging was performed.46,64Also, except for the clavicle, high proportions of conversion to operative management were observed for upper extremity fractures. However, these findings were based on case series in which it wasn’t always clear whether included patients had no clinical indication for post-treatment imaging, such as persistent pain. One RCT conducted in patients with distal radius fractures did not show a difference in complications or functional recovery in patients with a reduced imaging approach.82 We found consistent evidence for routine follow-up X-rays and consultations for lower extremity fractures (i.e., hip, femur, tibia, ankle, metatarsal); these practices were not associated with important changes in management or significant differences in complications or functional recovery when compared to selective use. However, apart from one RCT conducted in patients with ankle fractures84, these findings were mainly observed in case series. Globally, these results suggest that the number of follow-up X-rays should be questioned.