'Clinical scaphoid fracture': is it time to abolish this phrase?

Artificial intelligence for X-ray scaphoid fracture detection: a systematic review and diagnostic test accuracy meta-analysis

OBJECTIVES

Scaphoid fractures are usually diagnosed using X-rays, a low-sensitivity modality. Artificial intelligence (AI) using Convolutional Neural Networks (CNNs) has been explored for diagnosing scaphoid fractures in X-rays. The aim of this systematic review and meta-analysis is to evaluate the use of AI for detecting scaphoid fractures on X-rays and analyze its accuracy and usefulness.

MATERIALS AND METHODS

This study followed the guidelines of Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) and PRISMA-Diagnostic Test Accuracy. A literature search was conducted in the PubMed database for original articles published until July 2023. The risk of bias and applicability were evaluated using the QUADAS-2 tool. A bivariate diagnostic random-effects meta-analysis was conducted, and the results were analyzed using the Summary Receiver Operating Characteristic (SROC) curve.

RESULTS

Ten studies met the inclusion criteria and were all retrospective. The AI's diagnostic performance for detecting scaphoid fractures ranged from AUC 0.77 to 0.96. Seven studies were included in the meta-analysis, with a total of 3373 images. The meta-analysis pooled sensitivity and specificity were 0.80 and 0.89, respectively. The meta-analysis overall AUC was 0.88. The QUADAS-2 tool found high risk of bias and concerns about applicability in 9 out of 10 studies.

CONCLUSIONS

The current results of AI's diagnostic performance for detecting scaphoid fractures in X-rays show promise. The results show high overall sensitivity and specificity and a high SROC result. Further research is needed to compare AI's diagnostic performance to human diagnostic performance in a clinical setting.

CLINICAL RELEVANCE STATEMENT

Scaphoid fractures are prone to be missed secondary to assessment with a low sensitivity modality and a high occult fracture rate. AI systems can be beneficial for clinicians and radiologists to facilitate early diagnosis, and avoid missed injuries.

KEY POINTS

• Scaphoid fractures are common and some can be easily missed in X-rays. • Artificial intelligence (AI) systems demonstrate high diagnostic performance for the diagnosis of scaphoid fractures in X-rays. • AI systems can be beneficial in diagnosing both obvious and occult scaphoid fractures.

Deep Learning-Based Surgical Treatment Recommendation and Nonsurgical Prognosis Status Classification for Scaphoid Fractures by Automated X-ray Image Recognition

Biomedical information retrieval for diagnosis, treatment and prognosis has been studied for a long time. In particular, image recognition using deep learning has been shown to be very effective for cancers and diseases. In these fields, scaphoid fracture recognition is a hot topic because the appearance of scaphoid fractures is not easy to detect. Although there have been a number of recent studies on this topic, no studies focused their attention on surgical treatment recommendations and nonsurgical prognosis status classification. Indeed, a successful treatment recommendation will assist the doctor in selecting an effective treatment, and the prognosis status classification will help a radiologist recognize the image more efficiently. For these purposes, in this paper, we propose potential solutions through a comprehensive empirical study assessing the effectiveness of recent deep learning techniques on surgical treatment recommendation and nonsurgical prognosis status classification. In the proposed system, the scaphoid is firstly segmented from an unknown X-ray image. Next, for surgical treatment recommendation, the fractures are further filtered and recognized. According to the recognition result, the surgical treatment recommendation is generated. Finally, even without sufficient fracture information, the doctor can still make an effective decision to opt for surgery or not. Moreover, for nonsurgical patients, the current prognosis status of avascular necrosis, non-union and union can be classified. The related experimental results made using a real dataset reveal that the surgical treatment recommendation reached 80% and 86% in accuracy and AUC (Area Under the Curve), respectively, while the nonsurgical prognosis status classification reached 91% and 96%, respectively. Further, the methods using transfer learning and data augmentation can bring out obvious improvements, which, on average, reached 21.9%, 28.9% and 5.6%, 7.8% for surgical treatment recommendations and nonsurgical prognosis image classification, respectively. Based on the experimental results, the recommended methods in this paper are DenseNet169 and ResNet50 for surgical treatment recommendation and nonsurgical prognosis status classification, respectively. We believe that this paper can provide an important reference for future research on surgical treatment recommendation and nonsurgical prognosis classification for scaphoid fractures.

3D printed guide-assisted percutaneous screw fixation for minimally displaced scaphoid waist fractures with delayed diagnosis or presentation

OBJECTIVES

To Investigate the value of 3D printed guide-assisted percutaneous management of minimally displaced scaphoid waist fractures(Herbert's B2) with delayed diagnosis or presentation.

METHODS

From October 2018 to February 2022, 10 patients with established delayed diagnoses and presentation of minimally displaced scaphoid waist fractures were treated with 3D printed guides assisted with percutaneous internal fixation without bone grafting. This technique was based on the patient's preoperative CT and imported into the software. Based on Boolean subtraction, the most centralized screw placement position was identified and a customized guide was produced. Intraoperative percutaneous insertion of the guide wire was assisted by the custom guide.

RESULTS

All 10 patients were successful in one attempt. The fractures healed at a mean of 7.7 weeks postoperatively (range 6-10 weeks). At a mean follow-up of 7.7 months (6-13 months), patients had excellent recovery of wrist function with minimal pain reduction. There were no major postoperative complications and the patients all returned to their previous activities before the injury.

CONCLUSIONS

Percutaneous internal fixation based on 3D printed guides is a safe and effective technique for delayed diagnosis or presentation of patients with minimally displaced fractures of the scaphoid waist. This method allows for easy insertion of screws and avoids multiple attempts.

Which clinical features best predict occult scaphoid fractures? A systematic review of diagnostic test accuracy studies

BACKGROUND

Plain radiographs cannot identify all scaphoid fractures; thus ED patients with a clinical suspicion of scaphoid injury often undergo immobilisation despite normal imaging. This study determined (1) the prevalence of scaphoid fracture among patients with a clinical suspicion of scaphoid injury with normal radiographs and (2) whether clinical features can identify patients that do not require immobilisation and further imaging.

METHODS

This systematic review of diagnostic test accuracy studies included all study designs that evaluated predictors of scaphoid fracture among patients with normal initial radiographs. Quality assessment was undertaken using the Quality Assessment of Diagnostic Accuracy Studies 2 tool. Meta-analyses included all studies.

RESULTS

Eight studies reported data on 1685 wrist injuries. The prevalence of scaphoid fracture despite normal radiographs was 9.0%. Most studies were at overall low risk of bias but two were at unclear risk; all eight were at low risk for applicability concerns. The most accurate clinical predictors of occult scaphoid fracture were pain when the examiner moved the wrist from a pronated to a supinated position against resistance (sensitivity 100%, specificity 97.9%, LR+ 45.0, 95% CI 6.5 to 312.5), supination strength <10% of contralateral side (sensitivity 84.6%, specificity 76.9%, LR+ 3.7, 95% CI 2.2 to 6.1), pain on ulnar deviation (sensitivity 55.2%, specificity 76.4%, LR+ 2.3, 95% CI 1.8 to 3.0) and pronation strength <10% of contralateral side (sensitivity 69.2%, specificity 64.6%, LR+ 2.0, 95% CI 1.2 to 3.2). Absence of anatomical snuffbox tenderness significantly reduced the likelihood of an occult scaphoid fracture (sensitivity 92.1%, specificity 48.4%, LR- 0.2, 95% CI 0.0 to 0.7).

CONCLUSION

No single feature satisfactorily excludes an occult scaphoid fracture. Further work should explore whether a combination of clinical features, possibly in conjunction with injury characteristics (such as mechanism) and a normal initial radiograph might exclude fracture. Pain on supination against resistance would benefit from external validation.

TRIAL REGISTRATION NUMBER

CRD42021290224.

Diagnostic Performance and Advanced Imaging Reduction With Digital Tomosynthesis in Scaphoid Fracture Management

BACKGROUND

Management of scaphoid fractures often requires advanced imaging to achieve accurate diagnoses and appropriate treatment. Digital tomosynthesis (DTS) is a cross-sectional imaging modality that may be used to substitute magnetic resonance imaging or computed tomographic scans. The purpose of this study is to: (1) determine the diagnostic accuracy of DTS in occult scaphoid fractures; and (2) report on the reduction of other advanced imaging when using DTS.

METHODS

From May 2014 to October 2017, the charts of all patients who underwent scaphoid tomogram were retrospectively reviewed. The diagnostic accuracy of DTS for occult fracture was compared with 2-week follow-up plain films. To measure the reduction in utilization of advanced imaging, it was determined whether DTS eliminated the need for advanced imaging by providing adequate information regarding the clinical question.

RESULTS

A total of 78 patients underwent scaphoid tomography in this time frame: 39 for occult fracture, 33 for fracture union, 5 for fracture morphology, and 1 for hardware positioning. For the detection of occult fracture, DTS had a sensitivity of 100%, specificity of 83%, positive predictive value of 64%, and negative predictive value of 100%. Advanced imaging was not used in 35 of the remaining 39 patients based on the results obtained by DTS. In patients who did receive advanced imaging, 83% of tomograms provided conclusive diagnostic information.

CONCLUSIONS

Digital tomosynthesis increases the diagnostic sensitivity of occult scaphoid fractures, reducing unnecessary immobilization and advanced imaging. Digital tomosynthesis provides clinical detail beyond plain film, which reduces the need to obtain advanced imaging when assessing union, fracture pattern, and hardware placement.

Management and Outcomes of Clinical Scaphoid Fractures in Children

Background: Acute wrist trauma with clinical suspicion of a scaphoid fracture, but normal radiographs, is known as a clinical scaphoid fracture. Standard treatment involves immobilization and repeat radiographs in 10 to 14 days. When repeat radiographs are normal but a scaphoid fracture is still clinically suspected, the optimal management in children is unknown. This study retrospectively assessed the management and outcomes of pediatric patients diagnosed with clinical scaphoid fractures. Methods: A retrospective study was performed of all patients over a 2-year period treated for a clinical scaphoid fracture at a tertiary pediatric center. Patients were included if they had clinical signs of a scaphoid fracture and 2 negative x-rays 7 to 14 days apart postinjury. Results: Ninety-one patients with a mean age of 13.2 years (range: 7.8-17.7) were included. Sixteen patients (17.6%) underwent computed tomography (CT) or magnetic resonance imaging (MRI) at a mean time of 10.2 weeks postinjury. Five patients (5.5%) were diagnosed with a scaphoid fracture by x-ray or CT at an average of 4.5 weeks postinjury (range: 3-6). Six patients were diagnosed with other wrist fractures at a mean time postinjury of 3.1 (range: 3-6.5) weeks. Out of 195 total radiographs, the surgeon and radiologist disagreed on 59 (30.2%) images. No patients underwent surgery. Conclusions: Management of clinical scaphoid fractures at our institution was relatively uniform: nearly all patients were immobilized and less than 20% received advanced imaging. Our findings suggest a low but non-zero occult scaphoid fracture rate, discordance in radiologic interpretation, and lack of advanced imaging, providing an avenue for future prospective studies.

How to manage scaphoid waist fractures? Introduction of pragmatic imaging efficient pathway and literature review of the recent evidence

INTRODUCTION AND OBJECTIVE

The scaphoid bone sustains about 90 % of carpal bone fractures and is the second-highest bone at risk of post-traumatic osteonecrosis. Delayed diagnosis and treatment could lead to non-union and advanced carpal bones collapse. This study aimed to introduce an imaging efficient and practical scaphoid waist fracture management pathway (SWFMP) and measure its efficacy in clinical practice.

MATERIALS AND METHODS

The SWFMP was introduced in January 2020. Suspected occult fractures were approached by early orthopaedic clinical assessment and subsequent urgent MRI scan without repeating scaphoid X-rays. Scaphoid waist fractures displaced < 2 mm were treated with 8 weeks below elbow cast immobilization followed by CT scan if delayed union was suspected. Waist fractures displaced > 2 mm were managed with surgical fixation. Adult patients referred from the emergency department (ED) to the Virtual Fracture Clinic (VFC) with acute scaphoid injury from January 2019 to October 2019 (Pre SWFMP, n = 29), were identified and compared to those managed from January 2020 to October 2020 (Post SWFMP, n = 33).

RESULTS

Mean age was 37.9 (SD = 20.61) and 36.2 (SD = 17.06) years in the pre-SWFMP and post-SWFMP cohorts respectively. Fiften patients (51.7%) had the right side affected in the pre-SWFMP cohort and twenty-three patients (69.7%) in the post-SWFMP cohort. Scaphoid X-rays requested by ED have increased from 19 (65.5%) to 31 (94%) and repeated X-rays reduced from 17 (58.6%) to 10 (30.3%) after the introduction of the SWFMP. Mean wrist cast immobilization for patients without scaphoid fractures dropped from 16.9 days (SD = 5.57) to 3.6 days (SD = 6.24) after the SWFMP (p = 0.001). In the pre-SWFMP cohort, 24 patients had no fracture, 4 achieved full healing and 1 developed non-union. In the post-SWFMP cohort, 29 patients had no fracture, 1 achieved full union and delayed union was detected in 3 patients at 8 weeks.

CONCLUSIONS

The SWFMP has improved the clinical practice by reducing unnecessary ionizing radiation, unnecessary cast immobilization, and by using a timely fracture fixation intervention.

Development and Validation of a Deep Learning Model Using Convolutional Neural Networks to Identify Scaphoid Fractures in Radiographs

IMPORTANCE

Scaphoid fractures are the most common carpal fracture, but as many as 20% are not visible (ie, occult) in the initial injury radiograph; untreated scaphoid fractures can lead to degenerative wrist arthritis and debilitating pain, detrimentally affecting productivity and quality of life. Occult scaphoid fractures are among the primary causes of scaphoid nonunions, secondary to delayed diagnosis.

OBJECTIVE

To develop and validate a deep convolutional neural network (DCNN) that can reliably detect both apparent and occult scaphoid fractures from radiographic images.

DESIGN, SETTING, AND PARTICIPANTS

This diagnostic study used a radiographic data set compiled for all patients presenting to Chang Gung Memorial Hospital (Taipei, Taiwan) and Michigan Medicine (Ann Arbor) with possible scaphoid fractures between January 2001 and December 2019. This group was randomly split into training, validation, and test data sets. The images were passed through a detection model to crop around the scaphoid and were then used to train a DCNN model based on the EfficientNetB3 architecture to classify apparent and occult scaphoid fractures. Data analysis was conducted from January to October 2020.

EXPOSURES

A DCNN trained to discriminate radiographs with normal and fractured scaphoids.

MAIN OUTCOMES AND MEASURES

Area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity. Fracture localization was assessed using gradient-weighted class activation mapping.

RESULTS

Of the 11 838 included radiographs (4917 [41.5%] with scaphoid fracture; 6921 [58.5%] without scaphoid fracture), 8356 (70.6%) were used for training, 1177 (9.9%) for validation, and 2305 (19.5%) for testing. In the testing test, the first DCNN achieved an overall sensitivity and specificity of 87.1% (95% CI, 84.8%-89.2%) and 92.1% (95% CI, 90.6%-93.5%), respectively, with an AUROC of 0.955 in distinguishing scaphoid fractures from scaphoids without fracture. Gradient-weighted class activation mapping closely corresponded to visible fracture sites. The second DCNN achieved an overall sensitivity of 79.0% (95% CI, 70.6%-71.6%) and specificity of 71.6% (95% CI, 69.0%-74.1%) with an AUROC of 0.810 when examining negative cases from the first model. Two-stage examination identified 20 of 22 cases (90.9%) of occult fracture.

CONCLUSIONS AND RELEVANCE

In this study, DCNN models were trained to identify scaphoid fractures. This suggests that such models may be able to assist with radiographic detection of occult scaphoid fractures that are not visible to human observers and to reliably detect fractures of other small bones.

Has NICE guidance changed the management of the suspected scaphoid fracture: A survey of UK practice

INTRODUCTION

Despite scaphoid fractures being relatively uncommon pro-active treatment of suspected fractures has been seen as a risk management strategy. The poor positive predictive value of X-rays has led to published guidelines advocating MRI as a first-line or early imaging tool. It is unclear whether UK hospitals have been able to introduce early scanning and this national survey sought to establish the current management strategies for patients with a suspected scaphoid fracture.

METHOD

An electronic survey of UK emergency departments (ED) was conducted to establish the initial and follow up strategies for patients with negative imaging. Comparison of first and second-line imaging modalities was undertaken together with review of the clinical speciality responsible for ongoing management.

RESULTS

166 UK NHS Trusts were identified with emergency department facilities of which 66 (39.8%) responded. All sites perform an X-ray as the initial examination. For those with a negative examination ED follow up was the most common approach (54.6%), although many sites refer patients to other specialities including orthopaedics (39.4%) for follow up. The data demonstrated inconsistencies in the number of follow-up episodes and the different imaging investigations utilised. Frustration with the challenges presented by this patient cohort was evident.

CONCLUSION

The suspected scaphoid fracture represents an ongoing challenge to the NHS with many resource intensive pathways reliant on access to complex imaging investigations.

IMPLICATIONS FOR PRACTICE

Our study identified that UK Emergency Departments have limited early access to complex imaging for scanning of the scaphoid. A range of strategies are used for follow up of suspected scaphoid fractures and these are resource intensive. Overtreatment of patients with suspected scaphoid fracture is used as a risk management approach.

Diagnosis and Management of Acute Scaphoid Fractures

The scaphoid is the most commonly fractured carpal bone; despite its frequent injury, the diagnosis of fracture can be complicated by the presence of normal radiographs at the time of presentation. Clinical intuition can be increased by physical examination and immediately available modalities such as ultrasound within the emergency department. Definitive diagnosis should be made with computed tomography and magnetic resonance to verify the presence of displacement. This article provides an overview of the incidence and presentation of acute scaphoid fractures with a surgical focus on percutaneous dorsal screw fixation.

Why scaphoid fractures are missed. A review of 52 medical negligence cases

BACKGROUND

The "missed" scaphoid fracture is a common cause of litigation. This study investigates why a series of scaphoid fractures involved in medical negligence litigation were missed. It also assesses how many might have been detected if MRI had been used to assess all suspected scaphoid fractures with normal X-rays, as suggested by National Institute for Health and Care Excellence (NICE).

METHODS

Medical reports on 52 medical negligence cases of missed scaphoid fractures were reviewed. Complete sets of hospital and GP records and X-rays had been reviewed in every case.

RESULTS

The recorded mechanism of injury was consistent with a scaphoid fracture in 41 (79%) cases. There was no record of an examination for scaphoid tenderness in 37 (71%) cases. Scaphoid tenderness was assessed and not found in 10 of 15 cases. No X-rays were obtained in 13 (25%) cases, and a scaphoid fracture was only suspected, resulting in scaphoid series X-rays being performed, in only 3 (6%) cases. In only 5 (10%) cases was there a record of the patient being advised to return if the wrist remained painful.

CONCLUSIONS

Most of these scaphoid fractures were missed due to failure to consider the possibility of a scaphoid fracture and search for clinical signs of this injury. Some were missed due to failure to detect (or absence of) tenderness over the scaphoid bone. As a scaphoid fracture was never considered, or excluded by clinical examination, in 49 of the 52 cases, a policy of obtaining MRI for all suspected scaphoid fractures would only have detected three of 52 (6%) fractures. Improved awareness through better education is required to reduce the number of missed scaphoid fractures.

Early MRI versus conventional management in the detection of occult scaphoid fractures: what does it really cost? A rural pilot study

INTRODUCTION

To compare the cost-effectiveness and patient impact between acute magnetic resonance imaging (MRI) management and conventional management in the diagnosis of occult scaphoid fractures in a rural setting.

METHODS

Consecutive patients presenting to a rural emergency department (ED) with a suspected scaphoid fracture were randomly assigned to either conventional management (6) or acute MRI management (10) (3 patients were excluded from the study analysis). All healthcare costs were compared between the two management groups and potential impacts on the patients' pain, mobility and lifestyle were also measured.

RESULTS

There were no significant differences between the two groups at baseline. There was one (10%) scaphoid fracture in the MRI group and none in the conventional group (P = 0.42). A larger proportion of other fractures were diagnosed in the MRI group (20% (2) vs. 16.7% (1), P = 0.87), as well as less clinic attendances (1 (0-2.25) vs. 4 (2.25-5)) and diagnostic services (1 (1-1.25) vs. 2 (1-3)). Median management costs were $485.05 (AUD) (MRI) and $486.90 (AUD) (conventional). The MRI group had better pain and satisfaction scores as well as less time of immobilisation, treatment and time off work.

CONCLUSION

MRI dramatically reduces the amount of unnecessary immobilisation, time of treatment and healthcare usage in a rural setting. The two protocols are suggested to be equivalent financially. When potential societal costs, the amount of unnecessary immobilisation, low prevalence of true fractures and patient satisfaction are considered, acute MRI should be the management technique of choice. Further studies are still required to assess the best method for managing bone bruise within the scaphoid.

Scaphoid fractures: a higher order analysis of clinical tests and application of clinical reasoning strategies

INTRODUCTION

The purpose of this study was to identify clinical tests for scaphoid fractures and, using a higher order analysis, to determine their diagnostic accuracy.

METHODS

A literature review of the databases CINAHL, Embase, Medline and PUBMED from 1980 to September 30, 2011 was conducted to obtain applicable literature on clinical tests used in identifying scaphoid fractures. Methodological quality was determined using the criteria for validity suggested by Sackett et al (2000). Using a random effects model, pooled positive likelihood ratios (PPLR) were established for any test evaluated in at least 3 published studies. Moderator analyses provided insight into heterogeneity of results.

RESULTS

Higher order analysis indicated that the scaphoid compression test, anatomical snuffbox tenderness and scaphoid tubercle tenderness demonstrated statistically significant ability to identify scaphoid fractures with PPLR of 2.37 (1.27-4.41), 1.52 (1.12-2.06) and 1.67 (1.33-2.09) respectively. Descriptive factors (gender and mechanism of injury) were also identified but did not demonstrate significant diagnostic ability. Pooled data revealed the existence of heterogeneity for the three clinical tests and descriptive factors, which could not be easily explained.

CONCLUSION

Three clinical tests with statistically significant diagnostic validity were identified. In isolation, the clinical significance of each is questionable. Further studies with description of sample characteristics, blinded assessments, and agreement on a reference standard are recommended.

The impact of immediate reporting on interpretive discrepancies and patient referral pathways within the emergency department: a randomised controlled trial

OBJECTIVE

To determine whether an immediate reporting service for musculoskeletal trauma reduces interpretation errors and positively impacts on patient referral pathways.

METHODS

A pragmatic multicentre randomised controlled trial was undertaken. 1502 patients were recruited and randomly assigned to an immediate or delayed reporting arm and treated according to group assignment. Assessment was made of concordance in image interpretation between emergency department (ED) clinicians and radiology; discharge and referral pathways; and patient journey times.

RESULTS

1688 radiographic examinations were performed (1502 patients). 91 discordant interpretations were identified (n=91/1688; 5.4%) with a greater number of discordant interpretations noted in the delayed reporting arm (n=67/849, 7.9%). In the immediate reporting arm, the availability of a report reduced, but did not eliminate, discordance in interpretation (n=24/839, 2.9%). No significant difference in number of patients discharged, referred to hospital clinics or admitted was identified. However, patient ED recalls were significantly reduced (z=2.66; p=0.008) in the immediate reporting arm, as were the number of short-term inpatient bed days (5 days or less) (z=3.636; p<0.001). Patient journey time from ED arrival to discharge or admission was equivalent (z=0.79, p=0.432).

CONCLUSION

Immediate reporting significantly reduced ED interpretive errors and prevented errors that would require patient recall. However, immediate reporting did not eliminate ED interpretative errors or change the number of patients discharged, referred to hospital clinics or admitted overall.

ADVANCES IN KNOWLEDGE

This is the first study to consider the wider impact of immediate reporting on the ED patient pathway as a whole and hospital resource usage.