Citation: Mallee WH, Wang J, Poolman RW, et al. Computed tomography versus magnetic resonance imaging versus bone scintigraphy for clinically suspected scaphoid fractures in patients with negative plain radiographs. Cochrane Database Syst Rev. 2015;2015(6):CD010023. Published 2015 Jun 5. 

Link: https://pubmed.ncbi.nlm.nih.gov/26045406/

PDF

Article Summary: 

This article is a systematic review published in 2015 discussing the best imaging modalities for scaphoid fractures in patients with negative plain radiographs of the wrist. I chose this article because of a patient I saw in urgent care (H&P #3). The 25-year-old patient had multiple falls during his soccer game the day prior and reported right wrist/hand pain after his game. At urgent care the patient had anatomical snuffbox tenderness suggestive of a scaphoid fracture and was placed in a thumb spica splint. After sending the patient out for wrist x-rays, the radiologist called and confirmed that the patient had a scaphoid fracture. I recalled that plain radiographs may be negative for a scaphoid fracture for several weeks, so I wanted to learn more about the best imaging for such patients. This article is from within the last ten years and discusses and compares multiple imaging modalities for negative scaphoid fractures, which is why I included it for this summary. 

This review evaluated eleven prospective or retrospective studies that evaluated the accuracy of bone scintigraphy, computed tomography or magnetic resonance imaging for diagnosing scaphoid fractures. A total of 717 patients were included and indirect comparisons between modalities and participants were made. Twenty percent of scaphoid fractures have a normal radiograph, which can lead to non-union, avascular necrosis and carpal collapse if not diagnosed appropriately.  Thus, it is important to know the next best imaging to diagnose a scaphoid fracture. The data in this review found that bone scintigraphy was statistically the best modality for definitive diagnosis of a scaphoid fracture when x-ray was normal in comparison to CT or MRI. CT sensitivity ranged from 0.67 to 0.88 and specificity from 0.96 to 1.00. For MRI, sensitivity was 0.67 to 1.00 and specificity ranged from 0.89 to 1.00. Lastly, bone scintigraphy ranged from 0.95 to 1.00 and specificity from 0.52 to 1.00. While specificities were higher for CT and MRI, sensitivity, which produces a higher diagnostic accuracy, was significantly higher for bone scintigraphy than MRI or CT. Ultimately, this means that bone scintigraphy would reduce the risk of missing fractures but the low specificity would increase the number of patients who are unnecessarily treated in comparison to MRI and CT. 

This article did address the potential for bias and confounding variables due to the indirect comparison of imaging modalities. Additionally, while the results above currently suggest bone scintigraphy as the best method, it is the most invasive. It requires radioactive isotopes injected intravenously and can only be performed 72 hours after the injury in order to see the osteoblastic activity at the fracture location. Other potential issues with bone scintigraphy include the radiation from the imaging as well as the cost, which is similar to the cost of MRIs. 

Thus, while these results do provide some insight into what imaging modalities may be beneficial in suspected scaphoid fractures with negative plain radiographs, multiple other factors must be taken into consideration. Continuously understanding and learning about the best next-step imaging for suspected scaphoid fracture will aid in providing proper care.