Fractures of the scaphoid comprise 2% to 7% of all fractures and 51% to 90% of carpal fractures.1 The term “occult scaphoid fracture” refers to the 2% of fractures that plain radiographs are unable to detect in the acute setting. An undiagnosed scaphoid fracture presents a significant risk of subsequent non-union, given the susceptible vascular supply to this bone. Most suspected scaphoid injuries are first evaluated clinically and visualized on plain radiographs, allowing early initiation of treatment. However, when there is clinical evidence that raises the suspicion for a scaphoid fracture but no fracture is seen on initial radiographs, the possibility of an occult scaphoid fracture must be considered.
Diagnosis of occult scaphoid fractures remains a challenge, with various modalities advocated. Although there is evidence that early use of magnetic resonance imaging (MRI), bone scintigraphy, and computed tomography (CT) can accurately detect occult scaphoid fractures, these studies are time consuming and expensive and can be difficult to obtain because of scheduling or lack of availability. Digital tomography (DT) can assess for scaphoid fractures when there is high clinical suspicion but negative results on initial radiographs. This technique is a relatively new application of existing technology. It uses a single pass of a radiograph generator over a digital detector. Digital image slices between 1 and 5 mm wide are obtained within seconds. This imaging modality may improve conventional radiography of the scaphoid by rapidly producing morphologic, higher spatial resolution, CT-like images with minimal radiation exposure to the patient.
Digital tomography has been used to confirm the diagnosis of a suspected scaphoid fracture in a case report2 and has also been compared with repeat radiography at 2 weeks after trauma with favorable results.3 To the authors' knowledge, however, no literature exists objectively comparing the use of DT with the use of MRI in the analysis of carpal fractures. The authors conducted a prospective study to determine the usefulness of DT in detecting acute occult scaphoid fractures and to compare this modality in a blinded fashion with MRI, a well-established but much more expensive technique. The potential impact of this study would be to validate DT as a new clinical application of a currently available technology in an effort to improve diagnosis and management while minimizing radiation exposure.
Materials and Methods
Using an institutional review board–approved protocol, the authors enrolled a consecutive series of 39 patients between January 1, 2014, and July 1, 2014, for whom they had a high suspicion of occult scaphoid fracture. All patients were initially evaluated by a staff orthopedic surgeon and enrolled within 96 hours of the acute carpal trauma. Only patients with a mechanism of injury and symptoms consistent with an acute scaphoid fracture, including radial-sided pain and swelling at the base of the thumb, who did not have radiographic evidence of scaphoid fracture were included in the study. Once it was determined that they met this inclusion criterion, patients gave informed consent. Patients younger than 18 years (ie, being skeletal immaturity), pregnant patients, and those with comorbid conditions that precluded MRI were excluded. Both a digital tomogram and a wrist MRI were obtained for all patients within 1 to 3 days of presentation (Figure).
Plain wrist radiograph of a 24-year-old man with radial-sided wrist pain following a fall (A). Digital tomogram of the same patient showing scaphoid waist fracture (B).
All patients in the study were treated with a standard protocol, including thumb spica immobilization of the wrist. Evaluation at clinical follow-up was 10 to 14 days after initial presentation. Further routine clinical examinations and imaging ensued at appropriate intervals determined by the clinician (Table 1).
Application of Treatment Protocol Over Time
Digital tomography was performed with the VolumeRAD device (General Electric, Pewaukee, Wisconsin). For comparison, the radiation from a single chest radiograph exposure is comparable to that from 135 digital tomograms of the wrist. The authors used 2-mm collimated coronal images. Magnetic resonance imaging of the affected wrist, consisting of fat-suppressed coronal fast spin echo T2-weighted and spin echo T1-weighted images, was also performed.
Clinical Outcome and Statistical Techniques
Tomograms were analyzed in a blinded fashion by a staff radiologist (K.K.L.) without knowledge of the patients' MRI results. Repeat radiographs and clinical follow-up were used as a control to compare the efficacy of both digital tomogram and MRI. Data analysis was conducted by a biostatistician (S.S.S.). All analyses for significant differences were conducted using SPSS Statistics for Windows version 17.0 or higher software (IBM Corp, Armonk, New York) and confirmed by SAS software (SAS Institute, Cary, North Carolina). Analyses for statistical measures of performance (eg, sensitivity, positive predictive value) were conducted using Excel (Microsoft Corp, Redmond, Washington), with significance levels confirmed using SPSS software.
A total of 40 extremities in 39 patients were included in the study. The average age of the patients was 25.7 years. All patients underwent all diagnostic techniques. Digital tomogram had positive results for fracture in 4 (10%) of the 40 extremities at presentation, whereas MRI had positive results for 8 (20%) of the 40 extremities. The diagnosis rate was significantly higher for MRI (t=2.108, P=.041). On repeat radiographs at clinical follow-up, 6 (15%) of the 40 scaphoids were determined to be fractured. This resulted in a calculated sensitivity of 67% vs 100% (t=4.439, P=.0001) and positive predictive value of 100% vs 75% (t=3.651, P=.0008) for DT and MRI, respectively, when compared with follow-up radiographic findings (Table 2).
Sensitivity and Positive Predictive Value by Modality
The challenge of accurately diagnosing an occult scaphoid fracture has been well described. Failure to identify scaphoid fractures and initiate early treatment is associated with considerable morbidity. However, traditional management consisting of 2 weeks of immobilization until follow-up repeat radiographs when clinical suspicion for fracture is high is often inconvenient for patients. Pillai et al4 found that, in 90 patients with clinical signs suggestive of scaphoid injury, the incidence of true fracture was 6.66%, which is similar to the current findings. Thus, most patients who are treated for a suspected scaphoid fracture are unnecessarily immobilized, having a considerable loss of productivity and an increase in health care costs. Alternatively, if advanced imaging could be used to establish a definitive and timely diagnosis, patients could avoid unnecessary immobilization altogether.
Although the optimal algorithm for diagnosing occult scaphoid fractures has yet to be determined, various imaging modalities may complement radiographs in the detection of scaphoid injury. Magnetic resonance imaging has been shown to be very sensitive in the acute setting and is also capable of detecting ligamentous injury. In a study of 195 patients in whom scaphoid injury was clinically suspected but not indicated on initial radiographs, MRI detected fractures in 19% of these patients.5 Yin et al6 performed a meta-analysis of MRI vs bone scintigraphy, reporting equally high sensitivities for excluding scaphoid fracture but noting that MRI was more specific than bone scintigraphy (99% vs 89%). Other disadvantages of bone scintigraphy are the need for intravenous radioactive isotopes and a delay of 72 hours after injury before study interpretation. The use of MRI is further advocated by multiple studies on the cost-effectiveness of immediate MRI vs routine radiological follow-up of occult scaphoid fractures. These studies suggest that the 2 protocols are nearly equal from a financial standpoint in that patients are no longer unnecessarily immobilized in splints and thus do not lose productivity.7–9
Computed tomography can also be an effective diagnostic tool. However, unlike MRI, CT exposes patients to considerable radiation and has a high number of false-positive results. Thus, its utility has been suggested to be for ruling out an occult scaphoid fracture rather than for ruling in an occult scaphoid fracture.10 Furthermore, a CT scan may have limited diagnostic value when no fracture is present, especially if a soft tissue abnormality is present. Despite the common reference to MRI as a gold standard for diagnosing occult fractures, evidence suggests that its use is not without pitfalls. In a study of 100 patients presenting within 24 hours of injury, Beeres et al11 concluded that bone scintigraphy had superior sensitivity (100%) compared with early MRI (90%). Additionally, the expense and accessibility of MRI remain important considerations.
Digital tomography has been proposed as an imaging modality but has not been compared with MRI. The ability of DT to quickly produce high-resolution images without the radiation exposure of CT makes it a promising modality. In another study, DT was shown to be useful in the post-traumatic evaluation of consolidation or delayed union or nonunion, especially when osteosynthesis material had been used.2 This is an advantage over CT and MRI.
The current results indicate that DT detects most scaphoid fractures not seen on initial radiographs but may miss some entirely nondisplaced fractures, thereby making it less sensitive than MRI for detecting occult scaphoid fractures. Compared with repeat radiographs at clinical follow-up, DT missed identifying 2 of the 6 fractures. In the current series, there were 2 cases in which MRI indicated a fracture, a finding that was not corroborated by follow-up radiographs or clinical examination. The authors' assessment of these cases is that the signal seen in the scaphoid on the MRI likely represented a contusion without a true fracture. In the absence of a perfect imaging modality, it would seem prudent to select a test such as MRI with a maximum sensitivity (thus not missing any acute scaphoid fractures) while accepting less than perfect specificity, perhaps unnecessarily immobilizing a small fraction (5.9% in this study) of patients. Digital tomography, on the other hand, would have missed 2 fractures (33% of the total scaphoid fractures) in the current series, although it would not have led to unnecessary immobilization of any patients. Therefore, DT may be considered superior to MRI strictly in comparing cost, accessibility, radiation exposure, and the significantly fewer occurrences of false alarms due to over-sensitivity but not in the overall clinical management of scaphoid fractures. A limitation of these data is the lack of documented long-term follow-up, although such outcome measures are slightly out of the scope of this study.
The current results also serve to complement research findings in the military patient population. A previous study of scaphoid fracture incidence performed at the authors' institution reported 43 scaphoid fractures per 100,000 population.12 A study of military personnel treated at Beaumont Army Medical Center in Texas reported a 4-fold increase in incidence compared with the general population, with white men 20 to 24 years old being at greatest risk.13 These studies suggest that early, cost-effective, and readily available identification methods, which lead to decreased scaphoid nonunion and carpal morbidity, are even more crucial for the military active duty patient population, given its proportionally higher incidence of scaphoid injury. Other benefits include reductions in clinical appointments, unnecessary extremity immobilization, radiation exposure, and loss of productivity.
The authors have reported that DT provides imaging comparable to that of the current standard of care techniques in the detection of occult scaphoid fractures with minimal additional risk to patients. Neither DT nor MRI is a perfect imaging modality: DT missed 2 of the 6 occult fractures detected by plain radiographs at 2-week follow-up, and MRI misinterpreted 2 cases of contusion as fracture. This study showed that, although DT cannot replace MRI as the gold standard for the diagnosis of occult scaphoid fracture, DT is a cost-effective and readily available modality that is an effective adjunct to standard practice. As a diagnostic tool, DT may increase the efficiency of diagnosis, minimize unnecessary extremity immobilization, and possibly reduce health care costs in both the military and the general patient population.
- Tiel-van Buul MM, Roolker W, Broekhuizen AH, Van Beek EJ. The diagnostic management of suspected scaphoid fracture. Injury. 1997; 28(1):1–8. doi:10.1016/S0020-1383(96)00127-1 [CrossRef]
- Mermuys K, Vanslambrouck K, Goubau J, Steyaert L, Casselman JW. Use of digital tomosynthesis: case report of a suspected scaphoid fracture and technique. Skeletal Radiol. 2008; 37(6):569–572. doi:10.1007/s00256-008-0470-3 [CrossRef]
- Geijer M, Borjesson AM, Gothlin JH. Clinical utility of tomosynthesis in suspected scaphoid fracture: a pilot study. Skeletal Radiol. 2011; 40(7):863–867. doi:10.1007/s00256-010-1049-3 [CrossRef]
- Pillai A, Manav J. Management of clinical fractures of the scaphoid: results of an audit and literature review. Eur J Emerg Med. 2005; 12(2):47–51. doi:10.1097/00063110-200504000-00002 [CrossRef]
- Brydie A, Raby N. Early MRI in the management of clinical scaphoid fracture. Br J Radiol. 2003; 76(905):296–300. doi:10.1259/bjr/19790905 [CrossRef]
- Yin ZG, Zhang JB, Kan SL, Wang XG. Diagnosing suspected scaphoid fractures: a systematic review and meta-analysis. Clin Orthop Relat Res. 2010; 468(3):723–734. doi:10.1007/s11999-009-1081-6 [CrossRef]
- Raby N. Magnetic resonance imaging of suspected scaphoid fractures using a low field dedicated extremity MR system. Clin Radiol. 2001; 56(4):316–320. doi:10.1053/crad.2000.0657 [CrossRef]
- Dorsay TA, Majors NM, Helms CA. Cost-effectiveness of immediate MR imaging versus traditional follow-up for revealing radiographically occult scaphoid fractures. AJR Am J Roentgenol. 2001; 177(6):1257–1263. doi:10.2214/ajr.177.6.1771257 [CrossRef]
- Brooks S, Cicuttini FM, Lim S, Taylor D, Stuckey SL, Wluka AE. Cost effectiveness of adding magnetic resonance imaging to the usual management of suspected scaphoid fractures. Br J Sports Med. 2005; 39(2):75–79. doi:10.1136/bjsm.2003.007435 [CrossRef]
- Adey L, Souer JS, Lozano-Calderon S, Palmer W, Lee SG, Ring D. Computed tomography of suspected scaphoid fractures. J Hand Surg Am. 2007; 32(1):61–66. doi:10.1016/j.jhsa.2006.10.009 [CrossRef]
- Beeres FJ, Rhemrev SJ, Den Hollander P, et al. Early magnetic resonance imaging compared with bone scintigraphy in suspected scaphoid fractures. J Bone Joint Surg Br. 2008; 90(9):1205–1209. doi:10.1302/0301-620X.90B9.20341 [CrossRef]
- Burtis MT, Faillace J, Martin LF, Hermenau S. Scaphoid fracture detection in a military population: a standardized approach for medical referral. Mil Med. 2006; 171(5):404–408. doi:10.7205/MILMED.171.5.404 [CrossRef]
- Wolf JM, Dawson L, Mountcastle SB, Owens BD. The incidence of scaphoid fracture in a military population. Injury. 2009; 40(12):1316–1319. doi:10.1016/j.injury.2009.03.045 [CrossRef]
Application of Treatment Protocol Over Time
|Protocol Element||Time Period|
|Presentation||10- to 14-Day Follow-up||8 to 10 Weeks From Injury|
|Scaphoid view radiograph if not already available||Yes||Yes||Yes|
|Consent for study||Yes||NA||NA|
|Magnetic resonance imaging of affected wrist||Yesa||NA||NA|
Sensitivity and Positive Predictive Value by Modality
|Modality||No. of Positive Results for Scaphoid Injury||Sensitivitya||Positive Predictive Valuea|
|Initial medical history and physical examination||40||100%||15%|
|Repeat plain radiographs at 10 to 14 days||6||100%||100%|
|Magnetic resonance imaging||8||100%||75%|