a b s t r a c t

Objective: Many studies in the literature related to the investigation of the sensitivity and specificity of ultrasound examinations in lateral malleolar fractures is limited. The aim of this study is to investigate the sensitivity and specificity of ultrasound examinations performed by emergency physicians in fractures who are presented to the emergency department with blunt lateral malleolar trauma.

Method: Patients over 18 years of age who were admitted to the ED with lateral malleolar tenderness were en- rolled to this study with convenience sampling. Ultrasonographic examination was performed by emergency physicians. Following the ultrasound examination, a two-sided X-ray was performed. In the case of inconsistency between the US exam and the X-ray evaluated by the emergency physician, a CT was performed on the patients. The X-ray or CT imaging evaluation of an Orthopedic surgeon was accepted as the gold standard.

Results: A hundred-twenty patients were included in the study. Fractures in the Lateral malleolus were detected in 47 patients. The sensitivity of X-ray in the diagnosis of lateral malleolar fractures was 92.8%, (95% CI, 79.4-98.1) and the specificity was 100% (95% CI, 89.5-100), while the sensitivity of US exam was 100% (95% CI, 94.1-100), and the specificity was 93% (95% CI, 85-97.6). X-ray gave false negative results in 3 patients, whereas US gave false positive results in 5 patients.

Conclusion: In patients admitted to ED with lateral malleolus tenderness, the sensitivity of the ultrasound exam- ination performed by emergency physicians regarding diagnosis of lateral malleolar fracture is higher than X-ray.

(C) 2017

Introduction

Ankle sprains constitute approximately 5% of emergency depart- ment admission [1]. X-rays are performed on cases with lateral malleolus tenderness, under the Ottowa ankle rules, which are used to reduce the demand for radiography in emergency departments for cost effectivity. Nevertheless, less than 15% of the cases are fractured [2,3].

In the recent years, ultrasonography has started to be widely used for the diagnosis of bone fractures [4]. The advantages of ultrasonogra- phy are that it is inexpensive, does not cause radiation exposure, is easy to reproduce and provides easy imaging in a region where bone structure is superficial. Ultrasonography has been reported to have high accuracy in phalanx, ankle, metatarsal, metacarpal and distal

? This study presented as an oral presentation in 5th Eurasian Congress on Emergency Medicine, November 10 to 13, 2016, Antalya/Turkey.

* Corresponding author at: Dokuz Eylul University School of Medicine, Department of Emergency Medicine, Narlidere/Izmir, Turkey.

E-mail address: [email protected] (E. Aksay).

radius fractures [4-8]. In many of the studies it was observed that ultra- sonography detected some fractures that the X-rays had missed [8]. De- spite this, there are no studies focused on the use of ultrasonography in the diagnosis of Distal fibular fractures in patients with lateral malleolus tenderness.

We aim to investigate the accuracy of ultrasound examination per- formed by emergency physicians in fractures of adults who are present- ed to the emergency department with blunt lateral malleolar trauma.

Method

This prospective study was carried out at the Dokuz Eylul University, Department of Emergency Medicine after approval of the local ethics committee. Patients aged 18 years and older, admitted with lateral malleolar (distal 10 cm) tenderness after acute blunt trauma, between December 2014 and June 2015 were admitted to the study with conve- nience sampling. Demographic and clinal data were recorded on the study charts.

The US exam was performed by 3 emergency physicians and 2 emergency medicine residents. Before the study, 2 h didactic training

http://dx.doi.org/10.1016/j.ajem.2017.08.020

0735-6757/(C) 2017

P. Ozturk et al. / American Journal of Emergency Medicine 36 (2018) 362-365 363

was provided by a researcher who is experienced in fracture ultraso- nography to the other participating sonographers. Emergency medicine residents (sonographers) have not prior experience with musculoskel- etal ultrasound. Two of emergency physician had been conducted in one study with diagnostic musculoskeletal ultrasound, one of EP had been conducted in 5 studies on the diagnostic musculoskeletal ul- trasound before. All authors had been trained on ultrasound exam for emergency medicine practice for many years (FAST, vascular, echocardiograph etc.). The physicians participating in the study evaluat- ed 5 patients with lateral malleolus fractures and 5 patients without lat- eral malleolus fractures diagnosed by direct X-ray and then began to receiving patients to the study. The physicians were unblinded when performing US exams on the training session.

US exam were performed by using the Linear probe of the Mindray(R) M7 device. During imaging, the frequency of the device was set to 10 Mhz. Imaging was performed on the lateral, anterior and posterior surfaces of the lateral malleolus (10 cm distal of the ankle), transverse and longitudinal planes. US exam was completed before the X-ray so the sonographers were blinded from the X-ray result. A cortical irregu- larity on one or more plane considered as fracture for US exam.

Following the US examination, an anterior-posterior and lateral di- rect X-ray were performed on the patients via the Siemens(R) Optilix 150/50/50 HC-100. In the case of inconsistency between the US exam and the X-ray evaluated by the emergency physician, a CT was per- formed on the patients. The study protocol did not interfere with the clinical care of the patients.

An orthopedic surgeon who participated in the study and was un- aware of the US findings examined the X-ray images and if present CT images. In patients who underwent only X-ray imaging; the Orthopedic specialists’ evaluation of the X-ray was considered as gold standard, and if a CT was performed, the CT evaluation was considered as gold standard.

The following were considered as exclusion criteria for the study:

• • A patient refusing to be included in the study
  • Absence of the physician who should be performing the US exam in the emergency department
  • Inability to perform an US or X-ray on the patient for whatever reason
  • Loss of tissue integrity in the injury area
  • Trauma occurring one week prior to admission

Sample size analysis have not been conducted. Statistical analyses were performed by using SPSS 15.0 software (SPSS Inc., Chicago, IL). Qualitative data are presented as the number of observations and per- centage (%), while the quantitative data are presented as mean +- stan- dard deviation (SD) and (minimum-maximum). Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV), the likelihood ratio of the US exam and X-ray were calculated together with 95% confidence intervals. These investigations were carried out using the “Vassarstats” software. (http://vassarstars.net/)

Results

During the study period, 183 patients with a blunt ankle trauma who have tenderness to lateral malleol admitted to ED. Of those patients; 55 were excluded from the study due to the physician who would be performing the ultrasound being absent during the admission of the pa- tients to the ED, 4 patient were excluded due to the inability to obtain an X-ray image, 3 patients refused to be admitted to the study and a patient was excluded due to a trauma occurring one week prior to admission. A total of 120 (65.6%) patients participated in the study. Sixty-three (52.5%) were female and the mean age was 40.8 +- 19.3 years. Sixty- six patients (55%) exposure to the trauma on their right ankle.

One hundred and twenty patients underwent US exam and X-ray.

Table 1

Comparison of US exam with X-ray in fracture detection with the gold standard result

Imaging results		Gold standard
		Fracture present (n = 42)	No fracture (n = 78)
US exam	Fracture present (n = 47)	42	5
	No fracture (n = 73)	0	73
X-ray	Fracture present (n = 39)	39	0
	No fracture (n = 81)	3	78

Forty-seven patients (39.2%) were found to have fractures with US exam, and 39 (32.5%) fractures were found on X-ray. Forty-nine pa- tients (40.8%) who had inconsistencies between their X-ray and US exam underwent a CT.

A fracture was found in the CT of 3 patients who were not considered to have fractures in the X-ray but were fractured with the US exam (three false negative results for X-ray). There was no false positive result for X-ray. All fractures detected with US exam (sensitivity 100%), how- ever there were five cases with false positive US exam.

Comparison of US exam and X-ray results was shown in Table 1. The

sensitivity, specificity, negative predictive value, positive predictive value, and likelihood ratios of lateral malleolar fractures of the X-ray and US exam were shown in Table 2.

Discussion

We aimed to compare the accuracy of bedside US exam performed by emergency physicians with X-ray in the diagnosis of fractures in lat- eral malleol.

We found that the sensitivity of X-ray for detecting lateral malleolar fractures was 92.8% and the sensitivity for US exam was 100%. There was no false positive result with X-ray, whereas 5 false positive results were present with US exam (Specificity 100% for X-ray, 93% for US exam). Three patients who were not seen on X-ray but could be seen on CT were diagnosed with US exam. The USG, X-ray and CT images of a patient who was missed by X-ray are shown in Fig. 1.

We have 120 patients and five of them were with false positive re- sults. The order of these patients enrolled into study was 33, 36, 62, 80 and 97. Therefore, we though that we have sufficient experienced for US examination at the beginning of the study. Three false positive cases belong to emergency medicine residents who have not prior expe- rience with musculoskeletal ultrasound. However, two other false pos- itive results belong to emergency physician who is most experienced in musculoskeletal ultrasound. In one of this false positive case, the pa- tients was fractured his fibula and instrumentation were done eight years ago. Distal fibular plates were lead to false positive imaging on US exam (Fig. 2). Therefore, it should be considered that, in patients with previous lateral malleolar fracture is not suitable for US imaging.

There are many studies in the literature that have been using US exam for fracture diagnosis (Table 3). In most of these studies, the sen- sitivity and specificity of US exam were found to be similar with or su- perior from X-ray. Our study focused on only lateral malleolar trauma

Table 2

Sensitivity, specificity, NPD, PPD and LR (likelihood ratios) for detecting lateral malleolar fractures of the X-ray and US exam.

	X-ray	US exam
	%, (95% CI)	%, (95% CI)
Sensitivity	92.8, (79.4-98.1)	100, (89.5-100)
Specificity	100, (94.1-100)	93, (85-97.6)
NPV	96.2, (88.8-99)	100, (93.7-100)
PPV	100, (88.8-100)	89, (76.1-96)
LR (+)	?	8.4, (3.64-19.3)
LR (-)	0.03, (0.01-0.12)	0, (0 - ?)

NPV, Negative predictive value; PPV, positive predictive value; LR, likelihood ratio.

364 P. Ozturk et al. / American Journal of Emergency Medicine 36 (2018) 362-365

Fig. 1. X-rays (1A), coronal CT image (1B) and longitudinal US exam (1C) images of patients who was missed by X-ray.

differently from the other studies, because of lateral malleus is the most frequently fractured bone of ankle, US assessment of other bones such as a talus or navicular are difficult and complex trauma mechanism for medial malleolar fracture (higher risk of occurrence concomitant frac- ture of posterior malleolus or proximal fibula.

There are studies in the literature showing that the diagnostic sensitiv- ity of USG over ankle fractures is superior to X-ray. Simanovsky and et al. conducted a study on 20 patients; Ultrasound imaging was evaluated in pe- diatric patients with ankle injuries in whom X-rays were negative but clin- ical suspicion continued. X-rays obtained after 2 weeks were considered as gold standard in the study. A small metaphyseal cortical fracture was de- tected in the distal fibula in five of these patients with ultrasonography. A sub-cortical echogenic line was detected in the metaphysis of the fibula in one patient with ultrasound. No fracture was detected in the post- 2 week X-rays in any of the patients in whom no fractures were observed previously during the US exam. The US exam of the patients was performed by a pediatric radiologist comparatively with an intact ankle.

Atilla et al. investigated the diagnostic accuracy of US exam in frac- ture detection in patients with ankle sprains where the Ottowa imag- ing rules were indicated for X-ray. In the study, a 10 cm distal portion of the medial and lateral malleolus, a proximal-distal portion of the 5th metatarsal bone and the navicular bone were evaluated with an US exam. In the study, the X-ray and CT imaging interpretation of the orthopedists was considered as the gold standard. 246 patients were evaluated in this study and fractures were detected in 79 pa- tients according to gold standards. Lateral malleolus tenderness was present in 147 of the patients included in the study, the sensitivity of US exam in the diagnosis of lateral malleolar fractures was 86.8%, and the specificity was 97.3% [6]. The authors reported that there was a posterior malleolus fracture that was missed with X-ray, but that was diagnosed with US exam and CT.

Ekinci et al. investigated the accuracy of US exam in diagnosing fractures in ankle traumas of 131 patients who show indications for imaging in accordance with the Ottawa Ankle Rules. The ultraso- nographic imaging was performed by and emergency physician and the X-ray and CT imaging interpretation of the orthopedist was considered as the gold standard. The medial malleolus, lateral malleolus, talus, calcaneus, navicular, cuboid, cuneiform and meta- tarsal bones of the patients were evaluated with US exam. In 20% of the patients who were included in the study, fractures were de- tected with X-ray, of which 55% were observed in the lateral malleolus. In all 20 patients in which fractures were detected with X-ray, were also diagnosed with fractures with US exam. In one pa- tient, no fracture was detected with X-ray. However, a false positive fracture was detected in the ultrasound examination of the 1st metatarsal bone. In this study, the sensitivity of the ultrasound ex- amination regarding the recognition of ankle fractures was 100%, and the specificity was 99.1% [9].

Shojaee et al. studied the diagnostic sensitivity and specificity of US

exam in 141 patients with a suspicion of ankle fracture. The distal fibula, distal tibia, lateral malleolus, and medial malleolus were all evaluated in the patients who were included in the study. X-ray was considered as gold standard and fractures were detected in the X-rays of 102 of the 141 patients In this study, the sensitivity of the US exam regarding the determination of ankle fractures was 98.9%, and the specificity was 86.4% [10].

The studies investigating US accuracy for lateral malleolar fracture results with different diagnostic sensitivity. Atilla et al. reported their sensitivity as 86.8%, whereas Ekinci et al. and we found the sensitivity as 100%, Shojaee et al. found as 98.9%. The lower diagnostic sensitivity of US in the study of Atilla et al. may be due to difference of sonography device and physicians experience.

Fig. 2. In patients with false positive US Exam. Distal fibular plates were lead to false positive imaging (cortical irregularity) on US exam.

P. Ozturk et al. / American Journal of Emergency Medicine 36 (2018) 362-365 365

Table 3

Characteristics of studies on the sonographic diagnosis of fractures.

Author year	Localization	Population	Number of patients (fracture prevalence)	Sensitivity % (95% GA)	Specificity% (95% GA)	LR (+) (95% GA)	LR (-) (95% GA)
Atilla (2014)	Ankle	Adult	246	87.3	96.4	-	-
			(79)	(77.5-93.4)	(93.1%-98.2%)
Ekinci (2013)	Ankle	Adult	131	100	99.1	-	-
			(20)	(83.8-100)	(95-99.8)
Aksay (2013)	5th metacarpus	Adult	81	97.4	92.9	14	0.03
			(38)	(84.9-99.9)	(79.4-98.1)	(4.6-41)	(0.00-0.19)
Canagasabey (2010)	Ankle	Adult	110	90.9	90.9	10	0.1
			(11)	(65.7-98.3)	(88.1-91.7)	(5.5-11.9)	(0.02-0.4)
You (2010)	Sternum	Adult	36	100	100	-	-
			(24)	(90.3-100)	(90.3-100)
Ackermann (2009)	Forearm	Child	93	94	99	-	-
			(77)
Aksay (2016)	Proximal and intermediate phalanx	Adult	119	79.3	90	7.93	0.23
			(24)	(59.7-91.2)	(81.4-95)	(4.1-15)	(0.1-0.5)
Yesilaras (2015)	5th metacarpus	Adult	81	97.4	92.9	14	0.03
			(39)	(84.9-99.9)	(79.4-98.1)	(4.6-41)	(0-0.2)
Sivrikaya (2014)	Distal radius and ulna	Adult	90	100-89.5	88.2-94.6	13.1 (6.7-25.6)	0.02 (0-0.10)
			(79)
Shojaee (2016) [10]	Ankle	Adult	141	98.9	86.4	16	0.02
			(102)	(93.5-99.9)	(71.9-94.3)	(7.3-34.8)	(0.0003-0.182)

Limitations

The first limitation of our study is the patients enrolled into study with convenience sample. It was considered that this was a single center study over a short time period, which may not apply to other locations. The duration of the ultrasound examinations was not specified in our study. For this reason, no evidence could be provided that US exam is faster than an X-ray. The results of our study cannot be generalized for Penetrating injuries, children, and traumas older than one week. We performed a CT on patients with inconsistencies between their US exam and X-ray results, and the orthopedist made the decision in accor- dance with the CT result. If a CT had been done on all of the patients, it would be possible to show that both the US exam and the X-ray might have missed some fractures. While the orthopedist was completely blind regarding the examination findings of the patient, the physician performing the US exam (even if not examining the patient), may have realized the area with maximum sensitivity and may have investi- gated this are more carefully.

Conclusion

In patients in the adult age group admitted with lateral malleolus tenderness, bedside US exam performed by emergency physicians is more sensitive, but less specific than X-ray. US exam may be considered as a first line diagnostic tool in those patients. X-ray should be used only for confirmation in patients with positive US results.

References

1. Pijnenburg AM, Glas AS, Roos MAJ, Bogaard K, Lijmer JG, Bossuyt PMM, et al. Radiog- raphy in acute ankle injuries: the Ottawa ankle rules versus local diagnostic decision rules. Ann Emerg Med 2002;39:599-604.
2. Brooks SC, Potter BT, Rainey JB. Inversion injuries of the ankle: clinical assessment and radiographic review. Br Med J 1981;282:607-8.
3. Jenkin M, Sitler MR, Kelly JD. clinical usefulness of the Ottawa ankle rules for detect- ing fractures of the ankle and midfoot. J Athl Train 2010;45:480-2.
4. Aksay E, Yesilaras M, Kilic TY, Tur FC, Sever M, Kaya A. Sensitivity and specificity of bedside ultrasonography in the diagnosis of fractures of the fifth metacarpal. Emerg Med J 2015;32(3):221-5. http://dx.doi.org/10.1136/emermed-2013-202971.
5. Aksay E, Kilic TY, Yesilaras M, Tur FC, Sever M, Kalenderer O. Accuracy of bedside ul- trasonography for the diagnosis of finger fractures. Am J Emerg Med 2016;34(5): 809-12. http://dx.doi.org/10.1016/j.ajem.2016.01.010.
6. Atilla OD, Yesilaras M, Kilic TY, Tur FC, Reisoglu A, Sever M, et al. The accuracy of bed- side ultrasonography as a diagnostic tool for fractures in the ankle and foot. Acad Emerg Med 2014;21(9):1058-61. http://dx.doi.org/10.1111/acem.12467.
7. Yesilaras M, Aksay E, Atilla OD, Sever M, Kalenderer O. The accuracy of bedside ultra- sonography as a diagnostic tool for the Fifth metatarsal fractures. Am J Emerg Med 2014;32(2):171-4. http://dx.doi.org/10.1016/j.ajem.2013.11.009.
8. Sivrikaya S, Aksay E, Bayram B, Oray NC, Karakasli A, Altintas E. Emergency physi- cians performed Point-of-Care-Ultrasonography for detecting distal forearm frac- ture. Turk J Emerg Med 2016;19(16(3)):98-101.
9. Ekinci S, Polat O, Gunalp M, et al. The accuracy of ultrasound evaluation in foot and ankle trauma. Am J Emerg Med 2013;31:1551-5.
10. Shojaee M, Hakimzadeh F, Mohammadi P, et al. Screening characteristics of Ultraso- nography in detection of ankle fractures. Emerg (Tehran) 2016;11(4(4)):188-91.