Magnetic resonance imaging in the acute management of suspected scaphoid fractures: a review of the literature and assessment of treatment algorithm

Transcription

1 Page 1 of 6 Diagnosis Magnetic resonance imaging in the acute management of suspected scaphoid fractures: a review of the literature and assessment of treatment algorithm R Ganeshalingam 1*, K Eng 1,2, RS Page 1,2 Abstract Introduction The management of acute scaphoid fractures is optimised by early diagnosis and intervention. However, it can be difficult to diagnose those who present with normal initial X-rays. In this critical review, we looked at the use of early magnetic resonance imaging in clinically suspected but X-ray negative scaphoid fractures accompanied by data from our regional experience. Materials and methods A MEDLINE, EMBASE and Cochrane database search was performed assessing studies of acute wrist trauma. A total of 41 papers were reviewed. Results While previous literature on this topic is difficult to compare, overall, MRI offers improved sensitivity and specificity, detects additional wrist pathology, alters management in over 70% of cases and is a cost-effective tool. Discussion Our healthcare system utilises an acute wrist injury protocol whereby patients presenting with clinical suspicion of an acute scaphoid injury and a negative initial X-ray are referred for magnetic resonance imaging. We believe that this protocol avoids missed fractures (potentially minimising non-unions), unnecessary mobilisation and reduces overall costs to the healthcare system and community. *Corresponding author rekha4@gmail.com 1 Barwon Orthopaedic Research Unit, Barwon Health, Ryrie Street, Geelong, VIC 3220, Australia 2 School of Medicine, Deakin University, Barwon Health, Ryrie Street, Geelong, VIC 3220, Australia Conclusion More research is required to define what a fracture is on magnetic resonance imaging and how best to manage bone bruising with no cortical breach. Introduction Acute wrist injuries are a common presentation with clinical features of scaphoid fractures causing concern to both the patient and clinician even in the presence of normal X-rays. Scaphoid fractures may occur in as many as 16% of these cases 1, and the consequences of a missed diagnosis in this setting can be disabling and lead to medico-legal issues. However, overtreatment of patients results in increased healthcare costs and lost productivity. Further imaging of acute wrist injuries is controversial and needs to incorporate resource costs, radiation exposure and social and functional disability of the patient. Some healthcare systems offer a protocol for the patient with a symptomatic acutely injured wrist 2. Common protocols include immobilisation followed by repeat X-rays after 2 weeks 3, computer tomography (CT), magnetic resonance imaging (MRI) and nuclear imaging. Performing earlier, more accurate imaging prevents unnecessary immobilisation and delay in return to normal function. However, both CT and nuclear imaging come with increased radiation dose while MRI scanning is expensive. The aim of this review was to assess the impact of MRI scanning on acute wrist injuries with regards to fracture detection, detection of other injuries, alterations to management and cost effectiveness. Materials and methods A MEDLINE, EMBASE and Cochrane database search was performed using the terms wrist, scaphoid, wrist injury, acute and MRI. We included studies assessing patients with any acute wrist trauma and normal initial radiographs written in the English language. A total of 41 papers were reviewed. Results How accurate is MRI in diagnosing occult scaphoid fractures in those who have normal X-rays compared to other investigations? Forty-one studies were identified comparing the use of MRI to other imaging modalities in the detection of scaphoid fractures. The limitations of comparing previous research include: (1) varying patient inclusion criteria, from all acute wrist trauma with radial sided pain 4 to those with clinical features of scaphoid fracture but with a negative initial X-ray, (2) the lack of a gold standard to be used as a reference for such diagnostic studies 5,6 and (3) the definition of fracture on MRI, which may include or exclude isolated trabecular changes in the absence of a cortical breach (Table 1). Overall, MRI has been shown to detect scaphoid fractures missed by plain X-ray imaging Interobserver variation has been shown to be low in some studies 11 but only moderate in others 6. Compared to conventional radiography (X-ray) imaging, MRI has improved sensitivity (39% for X-ray vs 70% for MRI) and specificity (70% for X-ray vs 98% with MRI) in the detection of scaphoid fractures 4,10,12 14.

2 Page 2 of 6 Low et al. 10 showed that delayed follow-up radiographs alone in this patient population had poor sensitivity (9 49%), negative predictive value (30 40%) and poor inter-observer reliability. The study concluded that this is an inadequate diagnostic test Table 1 Reference Jorgsholm et al. 12 Lepisto et al. 20 Mack et al. 29 Mallee et al. 35 A C to exclude a scaphoid fracture in those with negative initial X-rays. MRI is comparable to, or out performs, other imaging modalities in diagnostic studies. A meta-analysis of 30 studies was performed by Yin et al. 5,15. This showed that follow-up Examples of defini on of fracture on MRI from different studies Criteria for fracture on magne c resonance imaging Cor cal and trabecular fractures present causing intramedullary hyperintensity on STIR as well as intra-medullary hypointensity on T1-weighted images extending to the cor ces T1 fracture as area of decreased signal intensity Cor cal fracture line Presence of a cor cal fracture line, a trabecular fracture line or a combina on of both Figure 1: Examples of other pathological findings seen on magnetic resonance imaging (MRI) included: (a) TFCC tear, (b) distal radius fracture, (c) bone bruising of the lunate and triquetrum and (d) Scapho-lunate ligament disruption, as indicated by the arrows. B D radiographs alone and CT had lower sensitivities (91.1 and 85.2%, respectively) than MRI (97.7%) and nuclear scanning (97.8%). Of note, however, the estimate of sensitivity for CT was thought to be imprecise due to lack of data. Specificities were found to be 93.5% for nuclear imaging, 99.8% for follow-up radiographs, 99.5% for CT and 99.8% for MRI. These results are comparable to other recent studies 1,16, where CT and MRI had similar diagnostic outcomes in the detection of scaphoid fractures with no significant difference in sensitivity and specificity. MRI detection of scaphoid fractures may, however, be affected by the low prevalence of the disease amongst patients presenting with negative initial X-rays 5. A randomised controlled trial 17 assessed patients presenting with acute wrist trauma (not specific to scaphoid injury features). Very few scaphoid fractures were found which makes analysis of its negative prediction value limited. Early MRI in all patients with acute wrist trauma is not recommended, however, it may be appropriate in those with clinical suspicion of scaphoid fracture. Clinical signs are integral to improving the detection of such injuries 3,18,19. What additional pathology is seen on MRI? MRI may identify additional wrist pathology including bony and soft tissue injuries (Figure 1). Previous studies have shown that four out of every five patients presenting with acute wrist trauma, but normal X-rays have a pathological finding on MRI 23. Jorgsholm et al. 4 diagnosed 224 fractures in the 196 wrists that were reviewed using MRI. These included scaphoid, distal radius fractures, triquetrum and capitate fractures. The most common multiple fracture combinations were scaphoid + distal radius and scaphoid + capitate fractures. The sensitivity of CT for visualisation of fractures other than scaphoid fractures was % and X-ray was <60%. Soft tissue injuries may include

3 Page 3 of 6 the triangular fibrocartilage complex (TFCC), the scapho-lunate ligament and other inter-carpal ligaments 7. However, the significance of some of these findings is unclear and must be related to the clinical presentation. What is bone bruising and how should it be treated? Bone bruising refers to a diffuse reduction in T1 signal with increase in T2 signal but with no linear component 24 and the management of such a finding is controversial. Some studies have allowed for no treatment of bone bruising 4 while others have recommended immobilisation for up to 2 weeks depending on the patients level of pain 23. Long-term outcomes were not, however, assessed. La Hei et al. 25 assessed 41 patients with acute scaphoid bone bruising and no fracture. These patients were immobilised for 6 weeks and the eight patients who remained symptomatic at the 3 month mark all showed improvement of bruising on repeat MRI. They concluded that bone bruising is a benign injury with predictable recovery and questioned the need for prolonged immobilisation. In contrast, Thavarajah et al. 24 raised concern about the potential progression of bruising to occult fracture. They prospectively assessed 170 scaphoid injuries, 50 of which had isolated bone bruising and no fracture. Following the theory that bone bruising represents marrow oedema and hence blood flow disruption, these patients were treated in scaphoid plaster for 8 weeks. There was one frank scaphoid fracture revealed on repeat MRI scanning at the end of these 8 weeks. It is unclear if this was misinterpreted on the MRI or whether it progressed. Neither study assessed the natural history of bone bruising in the absence of immobilisation. How does the MRI result alter patient treatment? The majority of studies with early MRI in suspected scaphoid injuries have shown a reduction in the incidence of missed fractures and that patients avoid unnecessary, prolonged immobilisation in those with a normal MRI. Mack et al. 14 prospectively reviewed 56 wrists with clinical suspicion of fracture but normal radiographs. Each wrist was examined with MRI in a mean of 6.6 days. The MRI data resulted in 66% of the cases having alterations to therapy, 22 (39%) wrists having a reduction in immobilisation time, 12 (21%) having a prolonged immobilisation and surgical intervention in 3 (5%). Seventy percent of the patients in this group were working prior to the injury. Similarly, Brydie et al. 22 showed alteration in management in 92% of their 195 patients as a result of MRI within 14 days of injury. Cost effectiveness While previously there has been concern about the cost effectiveness of performing a MRI scan on every patient with negative initial X-rays, the advent No wrist pathology identified Splint removed, Mobilisation encoraged of low field strength MRI 26 and more focused imaging has allowed for a reduction in costs and time constraints. Obtaining coronal T1-weighted and Short T1 Inversion Recovery (STIR) images now can take a similar time as would be required for scaphoid series plain X-rays (5 10 minutes) 21. Hansen et al. 27 assessed cost effectiveness by comparing 27 patients investigated with MRI within 1 week of injury to 27 patients managed with immobilisation and re-x-ray at 2 weeks post-injury. There was a statistically significant difference seen in those managed with MRI with savings of 2869 in non-hospital costs (loss of productivity). The difference was only statistically significant for manual workers. Our regional experience One hundred and ten wrists were reviewed in our centre for a suspected scaphoid fracture between August 2004 and March 2007 (see Figure 2 Patient presents with acute wrist injury and - Clinical suspicion of scaphoid fracture (e.g. anatomical snuffbox tenderness) - Normal wrist and scaphoid view xrays MRI performed either Acute scaphoid Sequence (T1 & T2 Fat Sat Coronal sequences) or Completed MRI wrist sequences Not significant Soft tissue injury Significant Fracture identified Plaster immobilization or referral for surgery as appropriate Figure 2: Management algorithm of patients presenting with clinical suspicion of scaphoid fracture at our hospital.

4 Page 4 of 6 A B C Figure 3: Presentation of a patient with an un-displaced scaphoid fracture, missed on plain X-ray (a) but detected on MRI (b) and (c), as indicated by the white arrows. for treatment algorithm). MRI was performed a mean of 6 days from injury (range: 1 21 days). Eighty-nine (81%) had a limited scaphoid trauma series MRI performed involving T1- and T2-fat saturated sequences (scan time of 5 minutes) and the remainder had a full six-sequence wrist MRI (scan time 25 minutes). Twenty-six patients (24%) had occult scaphoid fractures identified, defined as a bone lesion with oedema that breached the scaphoid cortex. Figure 3 shows an example of a scaphoid fracture initially missed on plain X-ray but detected in the T1- and T2-weighted images. Thirtysix patients (33%) had other fractures (see Table 2). Thirteen (12%) had significant soft tissue injuries (scapholunate ligament tears in 8, dorsal capsular tears in 3 and TFCC tears in 2) and 33 (30%) had normal results. 76% of patients with a normal initial X-ray had an alteration to their management as a result of the MRI findings. Patients with normal scans had their immobilisation removed. With regard to the 30% of the patients who avoided unnecessary X-rays, we estimated savings of $2400 (Aus) per person when taking into account the cost of MRI versus X-ray and loss of work time (see Table 3). Table 2 Injuries found on magne c resonance imaging in Barwon Health region Injury Number of pa ents (%) Scaphoid fracture 26 (24) Distal radial meta-physeal fracture 17 (15) Radial styloid 5 (4) Ulnar styloid 2 (1.8) Other carpal bone 7 (6) Metacarpal 5 (4) Scapholunate ligament tear 8 (7) Dorsal capsular tear 3 (2.7) Triangular fibro-car lage complex tear 2 (1.8) Table 3. Calcula on of savings due to early MRI for those pa ents with no wrist pathology but ini al clinical suspicion of scaphoid fracture Costs Average loss of earnings (2.5wks of wrist immobilisa on)* Tradi onal treatment algorithm $2, Treatment based on early MRI Savings per normal wrist with early MRI Cost of second X-ray + bone scan $ Cost of limited sequence MRI $ Total $2, *Based on average weekly wage of $ from Australian Bureau of Sta s cs for MRI, Magne c resonance imaging.

5 Page 5 of 6 Discussion The authors have referenced some of their own studies in this review. These referenced studies have been conducted in accordance with the declaration of Helsinki (1964) and the protocols of these studies have been approved by the relevant ethics committees related to the institution in which they were performed. All human subjects, in these referenced studies, gave informed consent to participate in these studies. Our healthcare system utilises an acute wrist injury protocol whereby patients presenting with clinical suspicion of an acute scaphoid injury and a negative initial X-ray are referred for MRI. We believe that this protocol avoids missed fractures (potentially minimising non-unions), unnecessary mobilisation and reduces overall costs to the healthcare system and community. From our own data, 76% of our patients had an alteration in their management as a result of this protocol, which is in keeping with other similar studies. Our own data support the use of MRI as a second line tool where plain XR is negative. It should be noted in our series, all the patients had tenderness in the anatomical snuff box, and we defined a fracture as including a cortical breach. The results from our local audit are comparable to other examples in the literature. The main difference we noted, however, was that amongst patients who received an MRI for scaphoid type pain, our centre had a greater prevalence of occult scaphoid fractures when compared to previous publications. This may have been due to improved clinical assessment with the majority of patients being reviewed by orthopaedic trainees or senior emergency doctors prior to ordering this investigation. MRI has improved sensitivity and specificity compared to conventional radiography and bone scintigraphy. Similar outcomes have been seen between MRI and CT scanning however MRI allows for detection and management of additional injuries and reduced radiation exposure. MRI is cost effective when compared to previous management protocols of further immobilisation and repeat X-ray. However, few studies have compared CT to MRI in terms of cost effectiveness. Jenkins et al. 1 showed reduced costs using CT imaging albeit with poorer diagnostic performance and they acknowledged the unknown long term cost benefit of identification of the additional injuries seen by MRI. Future studies need to look into the consequences of soft-tissue injuries identified by MRI and their economic burden. Also, the natural history of bone bruising is unknown and the optimal treatment method of such findings needs to be further assessed. Conclusion MRI in the setting of suspected scaphoid injuries and normal initial X-rays allows for optimisation of patient management and has been shown to be cost effective both in previous research and our own experience. We recommend the adoption of an MRI algorithm in the management of acute wrist injuries where the resources are available, based on a limited coronal trauma sequence. This enables the effective management of the spectrum of pathology seen, limits missed injuries, while providing appropriate early treatment for more significant pathology. References 1. Jenkins P, Slade K, Huntley JS, Robinson CM. A comparative analysis of the accuracy, diagnostic uncertainty and cost of imaging modalities in suspected scaphoid fractures. Injury Jan;39(7): Groves AM, Kayani I, Syed R, Hutton BF, Bearcroft PPW, Dixon AK, Ell PJ. An international survey of hospital practice in the imaging of acute scaphoid trauma. Am J Roentgenol Dec;187(6): Duckworth AD, Ring D, McQueen MM. Assessment of the suspected fracture of the scaphoid. J Bone & Joint Surg. (Br) Jun;93(6): Jorgsholm P, Thomsen NOB, Besjakov J, Abrahamsson SO, Bjorkman A. The benefit of magnetic resonance imaging for patients with posttraumatic radial wrist tenderness. J Hand Surg Jan;38(1): Yin ZG, Zhang JB, Kan SL, Wang XG. Diagnosing suspected scaphoid fractures. A systematic review and metaanalysis. Clin Orthop Relat Res Mar;468(3): De Zwart AD, Beeres FJ, Ring D, Kingma LM, Coerkamp EG, Meylaerts SA, Rhemrev SJ. MRI as a reference standard for suspected scaphoid fractures. Br J Radiol Aug;85(1016): Lepisto J, Mattila K, Nieminen S, Sattler B, Kormano M. Low field MRI and scaphoid fracture Aug;20(4): Hunter JC, Escobedo EM, Wilson AJ, Hanel DP, Zink-Brody GC, Mann FA. MR imaging of clinically suspected scaphoid fractures. AJR Am J Roentgenol May;168(5): Gaebler C, Kukla C, Breitenseher M, Trattnig S, Mittlboeck M, Vecsel V. Magnetic resonance imaging of occult scaphoid fractures. J Trauma Jul;41(1): Low G, Raby N. Can follow-up radiography for acute scaphoid fracture still be considered a valid investigation? Clin Radiol Oct;60(10): Beeres FJ, Hogervorst M, Kingma LM, Le Cessie S, Coerkamp EG, Rhemrev SJ. Observer variation in MRI for suspected scaphoid fractures. Br J Radiol Dec;81(972): Remplik P, Stabler A, Merl T, Roemer F, Bohndorf K. Diagnosis of acute fractures of the extremities: comparison of lowfield MRI and conventional radiography. Eur Radiol Apr;14(4): Lohman M, Kivisaari A, Vehmas T, Kinnunen J, Karaharju E, Kaukonen JP, Kivisaari L. MR imaging in suspected acute trauma of wrist bones. Acta Radiol. 199 Nov;40(6): Mack MG, Keim S, Balzer JO, Schwarz W, Hochmuth K, Windolf J, Vogl TJ. Clinical impact of MRI in acute wrist fractures. Eur Radiol Mar;13(3): Yin ZG, Zhang JB, Kan SL, Wang XG. Diagnosing accuracy of imaging modalities for suspected scaphoid fractures: meta-analysis combined with latent class analysis. J Bone Joint Surg Br Aug;94(8):

6 Page 6 of Mallee W, Doornberg JN, Ring D, van Dijk CN, Maas M, Goslings JC. Comparison of CT and MRI for diagnosis of suspected scaphoid fractures. J Bone Joint Surg Am Jan 5;93(1): Nikken JJ, Oei EH, Ginai AZ, Krestin GP, Verhaar JA, Van Vugt AB, Hunink MG. Acute wrist trauma: value of a short dedicated extremity MR Imaging examination in prediction of need for treatment. Radiology Jan;234(1): Powell JM, Lloyd GJ, Rintoul RF. New clinical test for fracture of the scaphoid. Can J Surg Jul;31(4): Freeland P. Scaphoid tubercle tenderness: a better indicator of scaphoid fractures? Arch Emerg Me Mar;6(1): Pierre-Jerome C, Moncayo V, Albastaki U, Terk MR. Multiple occult wrist bone injuries and joint effusions: prevalence and distribution on MRI. Emerg Radiol May;17(3): Khalid M, Jummani ZR, Kanagaraj K, Hussain A, Robinson D, Walker R. Role of MRI in the diagnosis of clinically suspected scaphoid fracture: analysis of 611 consecutive cases and literature review. Emerg Med J Apr;27(4): Page R, Partsalis T. MRI in the acute wrist injury. J Bone Joint Surg Br Orthopaedic Proceedings Mar;92B: Bergh TH, Lindau T, Bernardshaw SV, Behzadi M, Soldal LA, Steen K, Brudvik C. A new definition of wrist sprain necessary after findings in a prospective MRI study. Injury Oct;43(10): Thavarajah D, Syed T, Shah Y, Wetherill M. Does scaphoid bone bruising lead to occult fracture? A prospective study of 50 patients. Injury Nov;42(11): La Hei N, McFadyen I, Brock M, Field J. Scaphoid bone bruising probably not the precursor of asymptomatic non-union of the scaphoid. J Hand Surg Eur Vol Jun;32(3): Brydie A, Raby N. Early MRI in the management of clinical scaphoid fracture. Br J Radiol May;76(905): Hansen TB, Petersen RB, Barckman J, Uhre P, Larsen K. Cost-effectiveness of MRI in managing suspected scaphoid fractures. J Hand Surg Eur Vol Oct;34(5):