Diagnostic performance of ankle ultrasound for diagnosing anterior talofibular and calcaneofibular ligament injuries: a meta-analysis

Abstract

Background

Ankle ultrasound imaging could be an option with higher priority due to its lack of radiation, and cost- and time-effectiveness. However, previous studies regarding anterior tibiofibular ligament and calcaneofibular ligament injuries have shown varied results.

Purpose

To evaluate the diagnostic performance of ankle ultrasound for anterior tibiofibular ligament and calcaneofibular ligament injuries.

Material and Methods

PubMed and EMBASE databases were searched for diagnostic accuracy studies that used ultrasound for diagnosing anterior tibiofibular ligament and calcaneofibular ligament injuries. Bivariate and hierarchical summary receiver operating characteristic modeling were used to evaluate diagnostic performance. Subgroup analysis was performed using studies according to severity of the injury (complete and partial anterior tibiofibular ligament tear). We performed meta-regression analyses for heterogeneity exploration.

Results

Ten articles involving a total of 380 patients were included. For anterior tibiofibular ligament injury, the summary sensitivity, summary specificity, and area under the hierarchical summary receiver operating characteristic curve (AUC) were 0.99, 0.92, and 0.99, respectively. For calcaneofibular ligament injury, the summary sensitivity, summary specificity, and AUC were 0.95, 0.99, and 0.95, respectively. In subgroup analysis, for complete anterior tibiofibular ligament tear, the summary sensitivity, summary specificity, and AUC were 0.96, 0.82, and 0.96, respectively. For partial anterior tibiofibular ligament tear, the summary sensitivity, summary specificity, and AUC were 0.90, 0.82, and 0.93, respectively. Among the various potential covariates, proportion of anterior tibiofibular ligament tear, ultrasound interpreter, and reference standard were associated with specificity heterogeneity.

Conclusion

Ankle ultrasound demonstrates high diagnostic performance in the diagnosis of anterior tibiofibular ligament and calcaneofibular ligament injuries. We recommend ultrasound performed by a musculoskeletal radiologist as a first-line diagnostic tool to diagnose anterior tibiofibular ligament and calcaneofibular ligament injuries.

Keywords

Ankle ultrasound anterior talofibular ligament calcaneofibular ligament meta-analysis

Introduction

Lateral ankle sprains are the most common ankle joint injury, frequently occurring in daily and sports activities (1,2). It has been reported that >50% of adults may experience an ankle sprain (2). In lateral ankle sprains, the lateral collateral ligament (LCL) complex of the ankle, including the anterior talofibular ligament (ATFL), calcaneofibular ligament (CFL), posterior talofibular ligament, and anterior tibiofibular ligament can be injured. Among them, the ATFL and CFL are commonly damaged in ankle inversion because of their anatomical position (3).

ATFL and CFL injuries need adequate treatment plans (immobilization, rehabilitation, neuromuscular training) according to the injury grade; non-surgical treatment is usually chosen as a first-line option (4,5). If treatment after ATFL and CFL injury is inadequate, the injured joint is misused before healing, or rupture of any damaged ankle ligaments occurs, ankle instability may develop. Therefore, the early detection and accurate diagnosis of ATFL and CFL injury severity is crucial (3).

Clinical examination was initially used for the diagnosis of ATFL and CFL injuries in patients with ankle sprains. However, clinical examination alone may underdiagnose ATFL and CFL injuries (6); thus, imaging studies are commonly used. Ankle stress X-ray may visualize joint space, bony abnormalities including fractures, and abnormal soft tissue densities including swelling. It is especially useful as a primary screening tool for fractures, which may require timely intervention and which could be missed by ultrasound (US) (7). Computed tomography (CT) scan may visualize soft tissue more clearly than a plain X-ray, plus cross-sectional structures of bones (8,9). However, both procedures are hazardous because they use radiation sources and individuals must be exposed in a radioactive area. Magnetic resonance imaging (MRI) does not use radiation and is useful in imaging soft tissue, including ligament, tendon, and muscle, bleeding, and swelling; therefore, MRI may be used in a variety of soft tissue abnormalities, including ankle injuries. However, MRI is relatively expensive, time-consuming, and needs a fixed position to acquire images (10).

With this background, for the evaluation of the ATFL and CFL injuries, ankle US imaging could be an option with higher priority due to its lack of radiation, and cost- and time-effectiveness (10). However, previous US evaluation studies regarding ATFL and CFL injuries have shown varied results. Therefore, we believe that the performance of ankle US for diagnosing ATFL and CFL injuries needs further exploration and high-level evidence needs to be presented via quantitative synthesis of data from existing studies. Additionally, the pooling of results will be interesting because published studies have used different methodologies including injury severity (complete or partial tear) and reference standard (surgical finding or MRI finding).

This systematic review and meta-analysis aimed to evaluate the diagnostic performance of ankle US for diagnosing ATFL and CFL injuries. In addition, we performed a subgroup analysis to evaluate the diagnostic performance of ankle US according to severity of the injury (complete and partial tear).

Material and Methods

This meta-analysis followed the revised guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Diagnostic Accuracy Studies (PRISMA-DTA) statement (11).

Data sources

The PubMed and EMBASE databases were searched up to 2 January 2018 to identify English-language reports regarding US for diagnosing ATFL and CFL injuries. Search terms that were related to ‘ATFL,’ ‘CFL,’ and ‘ultrasound’ were combined with ‘sensitivity,’ ‘specificity,’ or ‘receiver operating characteristic’ as follows: ((‘anterior talofibular ligament’) OR (‘ATFL’) OR (‘calcaneofibular ligament’) OR (‘CFL’)) AND ((ultrasound) OR (ultrasonography) OR (sonogram) OR (sonography)) AND ((diagnosis) OR (accuracy) OR (sensitivity) OR (specificity) OR (receiver operating characteristic) OR (ROC curve)). The bibliographies of the identified articles were also screened to identify additional relevant studies. Two investigators screened the titles and abstracts for potential eligibility and disagreements were resolved through discussion.

Study selection

We included studies that fulfilled the following criteria: (i) patients with ankle injury; (ii) ankle US as the index test; (iii) use of surgical finding or MRI as the reference standard for confirmation of the ATFL and CFL ligaments tear; (iv) availability of sufficient information to reconstruct 2 × 2 contingency tables regarding sensitivity and specificity; and (v) original research article as the publication type with English language.

The exclusion criteria were as follows: (i) case report or case series; (ii) review articles, guidelines, consensus statements, letters, editorials, clinical trials, and conference abstracts; (iii) studies not pertaining to the field of interest; (iv) studies not performed on humans; (v) studies with insufficient data for 2 × 2 table; (vi) studies with only focused on diagnostic performance of indirect signs of the ankle ligaments tear such as joint effusion; and (vii) studies without use of surgical finding or MRI as the reference standard.

Data extraction and quality assessment

Two investigators independently extracted patient and study characteristics data. The same investigators evaluated methodological quality using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool (12). Inconsistencies between the reviewers were resolved through discussion.

A standardized form was used to extract data regarding: (i) patient characteristics (number of total patients, proportion of the ligament tear, mean age, age range, and sex); (ii) study characteristics (study location, publication year, study design, reference standard, and blinding to reference standard); and (iii) US characteristics (probe, technical parameters, and US performer). Study outcomes were also extracted to create 2 × 2 tables (i.e. true-positive, true-negative, false-positive, and false-negative results). The 2 × 2 tables were calculated using the Bayesian method if only sensitivity and specificity were presented for an eligible study. If two or more reviewers independently assessed the diagnostic accuracy, the result with the highest accuracy was extracted.

Data synthesis and analysis of diagnostic performance

Patient demographic characteristics and extracted covariates were summarized using standard descriptive statistics. Continuous variables are expressed as means and 95% confidence intervals (CIs), while categorical variables are expressed as frequencies or percentages unless stated otherwise.

We used a bivariate random-effects model for analyzing and pooling the diagnostic performance (sensitivity and specificity) measurements across studies. To derive summary estimates of diagnostic performance, we plotted estimates of the observed sensitivities and specificities for each test in forest plots and hierarchical summary receiver operating characteristic (HSROC) curves derived from individual study results (13 –15). These results were plotted using HSROC curves with 95% CIs and prediction regions.

Heterogeneity was determined using Cochran’s Q test (P < 0.05 indicated the presence of heterogeneity) and the I² test (0–40% = heterogeneity might not be present; 30–60% = moderate heterogeneity; 50–90% = substantial heterogeneity; and 75–100% = considerable heterogeneity) (16). When heterogeneity was noted, a ‘threshold effect’ was analyzed by visual assessment of the coupled forest plots of sensitivity and specificity. A meta-analysis of diagnostic test accuracy studies simultaneously evaluates a pair of outcomes (i.e. sensitivity and specificity). Sensitivity and specificity are commonly inversely correlated and influenced by the threshold (cut-off) value (13 –15). In addition, Spearman’s correlation coefficient between the sensitivity and false-positive rate was calculated to determine any threshold effect; a coefficient of >0.6 was considered to indicate a considerable threshold effect (17). We omitted Deeks’ funnel plot (18) of individual studies to check for publication bias according to the PRISMA-DTA.

Subgroup analysis

For detailed evaluation of the diagnostic performance of ankle US for diagnosing complete or partial ligament tear, we performed a subgroup analysis. We extracted the eligible studies in which ligament injury was classified as partial/complete or grade 2/3 for subgroup analysis. Grade 2 injuries were regarded as partial ligament tear and grade 3 injury was regarded as complete ligament tear (19).

Meta-regression analysis

Meta-regression analyses using several covariates were performed to explore the potential causes of heterogeneity: (i) patient characteristics (acute ankle injury versus chronic ankle instability with or without acute injury); (ii) total patients (≥50 versus <50); (iii) including pediatric patients (yes versus no); (iv) proportion of ATFL tear (≥50% versus <50%); (v) proportion of CFL tear (≥50% versus <50%); (vi) US performer (musculoskeletal radiologists versus others); and (vii) reference standard (operation versus MRI).

All statistical analyses were performed by one author who has three years of experience performing systematic reviews and meta-analyses. The statistical analyses were performed using the ‘midas’ and ‘metandi’ modules in Stata software (version 10.0; StataCorp LP, College Station, TX, USA) and the ‘mada’ package in R software (version 3.4.1; R Foundation for Statistical Computing, Vienna, Austria). Results were considered statistically significant at a P value < 0.05.

Results

Literature search

Fig. 1 shows a flow diagram summarizing the literature search. During the initial search, 227 studies were identified. After removing 34 duplicates, we reviewed 193 titles and abstracts, then excluded 166 studies for the following reasons: case reports/letters/editorials/conference abstracts (n = 24); review articles/guidelines/consensus statements (n = 24); not in the field of interest (n = 116); and not human study (n = 2). After reviewing the full text of 27 eligible articles, we excluded 17 for the following reasons: studies with insufficient data to build 2 × 2 tables (n = 11) (20 –30); studies focusing only on the diagnostic performance of indirect signs of ankle ligament tear (n = 1) (31); and studies without use of surgical finding or MRI as the reference standard (n = 5) (6,8,32 –34). Ultimately, 10 original research articles (19,35–43) including a total of 380 patients were included in the meta-analysis.

Fig. 1.

Study selection process for the meta-analysis.

Characteristics of the studies and included patients

The patient characteristics are summarized in Table 1. The sample size was in the range of 7–120 (age range = 12.1–34.0 years). The study and US characteristics are summarized in Table 2. All studies were prospective and single-centered, with blinding to the reference standard. Nine studies (19,35–40,42,43) performed consecutive enrollment and one study (41) performed case-control enrollment. All studies used linear array US transducer for evaluation of the ATFL and CFL.

Table 1.

The included patients’ demographic characteristics.

Article reference no.	Patient characteristics	Patients (n)	ATFL injury (%)	Complete ATFL tear (%)	CFL injury (%)	Complete CFL tear (%)	Mean age (years)	Age range (years)	Male (%)
36	Acute ankle injury	105	69	NR	55	NR	29.0	12–67	NR
42	Acute ankle injury	20	9	NR	4	NR	NR	NR	NR
41	CAI	17	14	8	0	NR	30.0	26–43	100
43	CAI	34	30	NR	N/A	NR	29.0	13–55	55.9
38	CAI	83	44	NR	N/A	NR	32.2	17–57	61.4
19	Acute ankle injury	30	22	10	N/A	2	NR	16–66	56.7
37	Acute ankle injury	65	32	NR	N/A	NR	34.0	18–72	63.1
35	CAI with/without acute injury	120	94	52	65	12	32.0	15–70	70.8
40	CAI with acute injury	85	65	NR	18	NR	27.3	18–40	89.4
39	Acute ankle injury	7	2	0	N/A	NR	12.1	5–17	NR

CAI, chronic ankle instability; ATFL, anterior talofibular ligament; CFL, calcaneofibular ligament; NR, not reported.

Table 2.

Study and US characteristics.

Article reference no.	Year	Locale	Study period	Study design	Reference standard	Time between injury and US	US model (manufacturer)	Frequency (MHz)	US performer
36	1993	Germany	NR	Prospective, consecutive	Operation	≤48 h	SLII (Siemens)	7.5	NR
42	1998	Germany	NR	Prospective, consecutive	MRI	NR	AU 530 (Esaote Biomedica)	13	NR
41	2008	United Kingdom	NR	Prospective, case-control	Operation	NR	ATL HDI 5000 (Phillips)	5–12	MSK radiologist
43	2010	Japan	Apr 2004–Dec 2006	Prospective, consecutive	Operation	4 weeks	Powervision (Toshiba)	9	OS, MSK radiologist
38	2012	China	Sep 2009–Sep 2010	Prospective, consecutive	Operation	NR	ATL HDI 5000 (Philips)	5–17	MSK radiologist
19	2012	Croatia	NR	Prospective, consecutive	MRI	1 week	Shimadzu 2200 (Shimadzu)	7–15	MSK radiologist
37	2013	Turkey	Jan 2011–Dec 2011	Prospective, consecutive	MRI	NR	Mindray DC-3 (Mindray)	7.5	EP
35	2014	China	Jan 2010–Nov 2012	Prospective, consecutive	Operation	>6 weeks	iU22 (Philips)	5–17	MSK radiologist
40	2017	South Korea	Oct 2016–Mar 2017	Prospective, consecutive	MRI	NR	Logiq E9 (GE)	12	EP, MSK radiologist
39	2018	Canada	Nov 2013–Feb 2014	Prospective, consecutive	MRI	≤72 h	z.one ultra (Zonare Medical Systems)	5–14	EP

US, ultrasound; MRI, magnetic resonance imaging; NR, not reported; OS, orthopedic surgeon; MSK, musculoskeletal; EP, emergency physician.

Quality assessment

Fig. 2 shows the risk of bias and applicability concerns for the 10 included studies. Overall, none of the studies were considered to be seriously flawed according to the QUADAS-2 tool. All the studies satisfied ≥4 of the seven items.

Fig. 2.

Grouped bar charts showing the risk of bias (left) and applicability concerns (right) for the 10 included studies using the Quality Assessment of Diagnostic Accuracy Studies-2 domains.

The risk of bias regarding the patient selection domain was considered high in one study (41) because this study performed case-control enrollment. Regarding the index test and reference standard domains, all studies were considered to have a low risk of bias. Regarding the flow and timing domain, seven studies (36 –40,42,43) had an unclear risk of bias because the mean interval between US and the reference standard was not reported. All studies exhibited low applicability to our research question in the patient selection, index test, and reference standard domains.

Overall diagnostic performance of US for diagnosing ATFL injuries

The 10 studies (19,35–43) had sensitivity values that were in the range of 0.50–1.00 and specificity values that were in the range of 0.25–1.00. The pooled sensitivity and specificity values were 0.99 (95% CI = 0.92–1.00) and 0.92 (95% CI = 0.76–0.97), respectively. The Q test revealed significant heterogeneity (Q = 11.012, P = 0.002), with considerable and substantial heterogeneity detected for sensitivity (I² = 85.01%) and specificity (I² = 80.52%), respectively. A threshold effect was observed in the coupled forest plot of sensitivity and specificity (Fig. 3) and in the correlation between sensitivity and the false-positive rate (0.110, 95% CI = −0.592 to 0.659). The area under the HSROC curve was 0.99 (95% CI = 0.98–1.00) (Fig. 4).

Fig. 3.

Coupled forest plots for the summary sensitivity and specificity of ankle ultrasound for anterior talofibular ligament injury. Dots in squares represent sensitivity and specificity. Horizontal lines represent the 95% confidence interval (CI) for each included study. The combined estimate (“Combined”) is based on the random-effects model and is indicated with diamonds. Corresponding heterogeneities (I²) with 95% CIs are provided at the bottom right corner: I²=100%×(Q−df)/Q, where Q is Cochran’s heterogeneity statistics and df is the degrees of freedom.

Fig. 4.

Hierarchical summary receiver operating characteristic (HSROC) curve for using ankle ultrasound for anterior talofibular ligament injury. The summary point (red box) indicates that the summary sensitivity was 0.99 and the summary specificity was 0.92. The 95% confidence region represents the 95% confidence intervals (CIs) of summary sensitivity and specificity, and the 95% prediction region represents the 95% CIs of sensitivity and specificity for each included study. The study estimates indicate the sensitivity and specificity estimated using the data from each study. The size of the marker is scaled according to the total number of patients in each study.

Overall diagnostic performance of US for diagnosing CFL injuries

Five studies (19,35,36,40,42) had sensitivity in the range of 0.94–1.00 and specificity in the range of 0.6–1.00. The pooled sensitivity and specificity values were 0.95 (95% CI = 0.89–0.98) and 0.99 (95% CI = 0.61–1.00), respectively. The Q test revealed significant heterogeneity (Q = 8.690, P = 0.006), with substantial and considerable heterogeneity detected for sensitivity (I² = 76.14%) and specificity (I² = 93.50%), respectively. A threshold effect was observed in the coupled forest plot of sensitivity and specificity (Fig. 5) and in the correlation between sensitivity and the false-positive rate (0.179, 95% CI = −0.514 to 0.792). The area under the HSROC curve (AUC) was 0.95 (95% CI = 0.93–0.97) (Fig. 6).

Fig. 5.

Coupled forest plots for the summary sensitivity and specificity of ankle ultrasound for calcaneofibular ligament injury.

Fig. 6.

Hierarchical summary receiver operating characteristic (HSROC) curve for using ankle ultrasound for calcaneofibular ligament injury. The summary point (red box) indicates that the summary sensitivity was 0.95 and the summary specificity was 0.99.

Subgroup analysis: diagnostic performance of ankle US according to severity of ATFL injury

Four studies (19,35,39,41) were included for subgroup analysis. In terms of complete ATFL tear, the pooled sensitivity and specificity were 0.96 (95% CI = 0.90–0.98) and 0.82 (95% CI = 0.72–0.89), respectively. The AUC was 0.96 (95% CI = 0.94–0.98). In terms of partial ATFL tear, the pooled sensitivity and specificity were 0.90 (95% CI = 0.83–0.94) and 0.82 (95% CI = 0.66–0.92), respectively. The AUC was 0.93 (95% CI = 0.90–0.95). We did not perform subgroup analysis with regard to CFL due to the small number of studies.

Meta-regression analysis

The results of the meta-regression analyses with regard to ATFL tear are summarized in Table 3. We did not perform meta-regression analysis with regard to CFL tear due to the small number of studies. The significant sources of heterogeneity in terms of sensitivity were proportion of ATFL tears (P < 0.01), US interpreter (P = 0.03), and reference standard (P < 0.01) with higher sensitivity reported in studies with relatively high proportion of the ATFL tear (≥50%), musculoskeletal radiologists, and surgical finding as reference standard than those with relatively low proportion of the ATFL tear (<50%), emergency physician, and MRI as reference standard. There was no significant source of heterogeneity in terms of specificity.

Table 3.

Meta-regression analyses for potential sources of heterogeneity in terms of anterior talofibular ligament (ATFL) tear.

Covariate	Studies (n)	Sensitivity (95% CI)	P value	Specificity (95% CI)	P value
Patient characteristic			0.84		0.99
Acute ankle injury	5	0.96 (0.88–1.00)		0.89 (0.73–1.00)
Chronic ankle instability	5	1.00 (0.98–1.00)		0.93 (0.82–1.00)
Patients (n)			0.22		0.06
≥50	5	0.99 (0.97–1.00)		0.97 (0.92–1.00)
<50	5	0.96 (0.89–1.00)		0.77 (0.56–0.97)
Including pediatric patients			0.15		0.28
Yes	6	0.99 (0.98–1.00)		0.83 (0.63–1.00)
No	3	0.99 (0.96–1.00)		0.99 (0.96–1.00)
ATFL tear (%)			<0.01		0.96
≥50	6	0.99 (0.98–1.00)		0.89 (0.75–1.00)
<50	4	0.89 (0.74–1.00)		0.95 (0.85–1.00)
CFL tear (%)			0.90		0.97
≥50	2	1.00 (0.99–1.00)		0.93 (0.83–1.00)
<50	3	0.99 (0.93–1.00)		0.94 (0.86–1.00)
US performer			0.03		0.66
MSK radiologist	5	0.99 (0.97–1.00)		0.92 (0.78–1.00)
Others	3	0.97 (0.90–1.00)		0.95 (0.85–1.00)
Reference standard			<0.01		0.81
Operation	5	1.00 (0.98–1.00)		0.88 (0.71–1.00)
MRI	5	0.96 (0.87–1.00)		0.94 (0.84–1.00)

Italic text indicates statistical significance (P < 0.05).

CI, confidence interval; No., number; ATFL, anterior talofibular ligament; CFL, calcaneofibular ligament; MSK, musculoskeletal; MRI, magnetic resonance imaging; US, ultrasound.

Discussion

The present meta-analysis revealed that ankle US was excellent for diagnosing ATFL (sensitivity = 99%, specificity = 92%) and CFL injuries (sensitivity = 95%, specificity = 99%). Considering these findings, ankle US is a useful imaging modality for diagnosing ATFL and CFL injuries. In subgroup analysis, diagnostic performance of ankle US for diagnosing complete and partial ATFL tears were excellent (complete ATFL tear, sensitivity = 96%, specificity = 82%; partial ATFL tear, sensitivity = 90%, specificity = 82%).

Traditionally, the initial assessment of the extent of ankle injuries has been done using radiography, including stress views (44,45). However, it is debatable about the value of stress radiographs, especially in the acute phase, because the findings are severely influenced by the radiographic technique, the amount of force applied to the joint, and patient cooperation. The latter depends on the amount of pain, refractory muscle spasm, and edema of the surrounding soft tissue (44). With this background, ankle US for diagnosing ATFL and CFL injuries has been investigated and recent advances in US equipment have considerably increased the diagnostic usefulness of this technique (7,46,47).

Our results may have important clinical implications. First, point-of-care US is increasingly being used to facilitate accurate and timely diagnoses and to plan treatment. Moreover, it may improve patient satisfaction because it can expedite clinical decision-making and is performed at the bedside (48 –50). Our results can be expanded on by making an indirect comparison between US and stress radiograph findings. A recent study (21) related to the comparison between ankle US and stress radiographs demonstrated that US is more accurate. In stress radiographs, anterior translation and talar tilt angle showed variable sensitivity (15.8–78.9%) and specificity (0–57.9%). These sensitivity and specificity values of stress radiographs are lower than those of US in our meta-analysis. Therefore, we speculate that US may be more useful than stress radiographs for the evaluation of ATFL and CFL injuries. Second, it is a dynamic study (and can include a stress test) and so is able to gauge how unstable the affected ankle is. Third, it can evaluate the entire portion of the ligament in longitudinal view (46). Third, it can evaluate the contralateral ankle and is thus able to compare whether the ligament or tendon is thinned/thickened or not. Also, the use of US may reduce the radiation dose. Therefore, on the basis of the ‘first do no harm’ principle and the ‘as low as reasonably achievable’ concept, we believe the use of US is warranted to reduce radiation exposure in patients with suspected ATFL and CFL injuries.

Our meta-regression analysis revealed that the US interpreter was a source of heterogeneity. In particular, the pooled sensitivity was higher in studies with musculoskeletal radiologists than in studies with emergency physicians. Thus, we recommend that musculoskeletal radiologists perform the initial US for diagnosing ATFL and CFL injuries. If clinical physicians, including emergency physicians, perform ankle US, we speculate that focused musculoskeletal US training is required. Actually, more recently, emergency US has been formally incorporated into emergency medicine fellowship training because US can show different diagnostic performance according to the operator’s proficiency (51).

This meta-analysis only examined studies where the diagnostic performance of US was based on conclusive cases, as the eligible studies did not include cases with equivocal or inconclusive findings. Moreover, almost all studies using US emphasized its diagnostic performance alone and did not compare it to other modalities. Thus, a comprehensive study that includes equivocal and inconclusive cases and other methodology may be needed to confirm the usefulness of ankle US as an initial diagnostic tool in routine clinical practice.

The present study has several limitations. The first limitation is the relatively small number of included studies. Nevertheless, we were able to draw several important conclusions regarding the diagnostic performance of US and related factors, which we believe provides a useful overview because we used broad search terms and only included easily accessible studies (published in English and available in the PubMed and EMBASE databases). The second limitation is that all included studies revealed positive results, and that fact could be attributed to publication bias. Although we omitted Deeks’ funnel plots according to the PRISMA-DTA guidelines, we observed a low probability of publication bias (overall, P = 0.75–0.90; subgroup analysis, P = 0.26–0.96), which suggests that this factor did not undermine our results. The third limitation is the methodological differences between the included studies, and the extensive meta-regression analysis revealed that these variables were also significant sources of heterogeneity. This methodological diversity might affect the pooled estimates, especially as the US technical parameters were not assessed in the meta-regression analysis because not all studies reported the values for gain, dynamic range, and mechanical index. Further studies with larger sample sizes are needed to determine the optimal parameters for ankle US.

In conclusion, ankle US, which is a non-invasive, radiation-free modality, demonstrates high diagnostic performance in the diagnosis of ATFL and CFL injuries. We recommend that US should be performed by musculoskeletal radiologists as a first-line diagnostic tool to more accurately diagnose ATFL and CFL injuries.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Czajka

Tran

Cai

, et al. Ankle sprains and instability. Med Clin North Am 2014; 98:313–329.

Doherty

Delahunt

Caulfield

, et al. The incidence and prevalence of ankle sprain injury: a systematic review and meta-analysis of prospective epidemiological studies. Sports Med 2014; 44:123–140.

Salat

Veljkovic

, et al. Imaging in foot and ankle instability. Foot Ankle Clin 2018; 23:499–522.

Al-Mohrej

Al-Kenani

NS.

Acute ankle sprain: conservative or surgical approach?

EFORT Open Rev 2016; 1:34–44.

Petersen

Rembitzki

Koppenburg

, et al. Treatment of acute ankle ligament injuries: a systematic review. Arch Orthop Trauma Surg 2013; 133:1129–1141.

Gremeaux

Coudreuse

Collado

, et al. Comparative study of clinical and ultrasonographic evaluation of lateral collateral ligament sprains of the ankle. J Sports Med Phys Fitness 2009; 49:285–291.

Guo

Zhu

, et al. Reliability of measurements on lateral ankle radiographs. BMC Musculoskelet Disord 2016; 17:297.

Guillodo

Varache

Saraux

Value of ultrasonography for detecting ligament damage in athletes with chronic ankle instability compared to computed arthrotomography.

Foot Ankle Spec 2010; 3:331–334.

Hong

Paeng

, et al. Comparison of SPECT/CT and MRI in diagnosing symptomatic lesions in ankle and foot pain patients: diagnostic performance and relation to lesion type. PLoS One 2015; 10:e0117583.

10.

Alves

Dong

Jacobson

, et al. Normal and injured ankle ligaments on ultrasonography with magnetic resonance imaging correlation. J Ultrasound Med. 2019; 38:513–528.

11.

McInnes

MDF

Moher

Thombs

, et al. Preferred Reporting Items for a Systematic Review and Meta-analysis of Diagnostic Test Accuracy Studies: The PRISMA-DTA Statement. JAMA 2018; 319:388–396.

12.

Whiting

Rutjes

Westwood

, et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med 2011; 155:529–536.

13.

Kim

Lee

Choi

, et al. Systematic review and meta-analysis of studies evaluating diagnostic test accuracy: a practical review for clinical researchers-part I. General guidance and tips. Korean J Radiol 2015; 16:1175–1187.

14.

Lee

Kim

Choi

, et al. Systematic review and meta-analysis of studies evaluating diagnostic test accuracy: a practical review for clinical researchers-part II. Statistical methods of meta-analysis. Korean J Radiol 2015; 16:1188–1196.

15.

Suh

Park

SH.

Successful publication of systematic review and meta-analysis of studies evaluating diagnostic test accuracy.

Korean J Radiol 2016; 17:5–6.

16.

Higgins

Green

Cochrane handbook for systematic reviews of interventions. Version 5.1.0. Oxford: The Cochrane Collaboration. Available at: http://handbook.cochrane.org/chapter_9/9_5_2_identifying_and_measuring_heterogeneity.htm.

17.

Deville

Buntinx

Bouter

, et al. Conducting systematic reviews of diagnostic studies: didactic guidelines. BMC Med Res Methodol 2002; 2:9.

18.

Deeks

Macaskill

Irwig

The performance of tests of publication bias and other sample size effects in systematic reviews of diagnostic test accuracy was assessed.

J Clin Epidemiol 2005; 58:882–893.

19.

Margetic

Pavic

Comparative assessment of the acute ankle injury by ultrasound and magnetic resonance. Coll Antropol 2012; 36:605–610.

20.

Campbell

Menz

Isaacs

Dynamic ankle ultrasonography. A new imaging technique for acute ankle ligament injuries.

Am J Sports Med 1994; 22:855–858.

21.

Cho

Lee

Song

, et al. Value of stress ultrasound for the diagnosis of chronic ankle instability compared to manual anterior drawer test, stress radiography, magnetic resonance imaging, and arthroscopy. Knee Surg Sports Traumatol Arthrosc 2016; 24:1022–1028.

22.

Croy

Saliba

, et al. Differences in lateral ankle laxity measured via stress ultrasonography in individuals with chronic ankle instability, ankle sprain copers, and healthy individuals. J Orthop Sports Phys Ther 2012; 42:593–600.

23.

Croy

Saliba

, et al. Talofibular interval changes after acute ankle sprain: a stress ultrasonography study of ankle laxity. J Sport Rehabil 2013; 22:257–263.

24.

Elkaïm

Thès

Lopes

, et al. Agreement between arthroscopic and imaging study findings in chronic anterior talo-fibular ligament injuries. Orthop Traumatol Surg Res 2018; 104:S213–S218.

25.

Hsu

Tsai

Chen

CPC

, et al. Ultrasonographic examination for inversion ankle sprains associated with osseous injuries. Am J Phys Med Rehabil 2006; 85:785–792.

26.

Liu

Gustavsen

Royer

, et al. Increased ligament thickness in previously sprained ankles as measured by musculoskeletal ultrasound. J Athl Train 2015; 50:193–198.

27.

Maeda

Takaoka

, et al. Sonographic findings of chondral avulsion fractures of the lateral ankle ligaments in children. J Ultrasound Med 2017; 36:421–432.

28.

Mizrahi

Nazarian

Parker

Evaluation of the Anterior Talofibular Ligament via Stress sonography in asymptomatic and symptomatic populations.

J Ultrasound Med 2018; 37:1957–1963.

29.

Szczepaniak

Ciszkowska-Lyson

Smigielski

, et al. Value of ultrasonography in assessment of recent injury of anterior talofibular ligament in children. J Ultrason 2015; 15:259–266.

30.

Yildizgoren

Velioglu

Demetgul

, et al. Assessment of the anterior talofibular ligament thickness in patients with chronic stroke: an ultrasonographic study. J Med Ultrasound 2017; 25:145–149.

31.

Guillodo

Riban

Guennoc

, et al. Usefulness of ultrasonographic detection of talocrural effusion in ankle sprains. J Ultrasound Med 2007; 26:831–836.

32.

Farley

Kuhns

Jacobson

, et al. Ultrasound examination of ankle injuries in children. J Pediatr Orthop 2001; 21:604–607.

33.

Lee

Park

Jegal

, et al. New method of diagnosis for chronic ankle instability: comparison of manual anterior drawer test, stress radiography and stress ultrasound. Knee Surg Sports Traumatol Arthrosc 2014; 22:1701–1707.

34.

Wiebking

Pacha

Jagodzinski

An accuracy evaluation of clinical, arthrometric, and stress-sonographic acute ankle instability examinations.

Foot Ankle Surg 2015; 21:42–48.

35.

Cheng

Cai

Wang

Value of ultrasonography for detecting chronic injury of the lateral ligaments of the ankle joint compared with ultrasonography findings.

Br J Radiol 2014; 87:20130406.

36.

Friedrich

Schnarkowski

Rubenacker

, et al. Ultrasonography of capsular morphology in normal and traumatic ankle joints. J Clin Ultrasound 1993; 21:179–187.

37.

Gün

Ünlüer

Vandenberk

, et al. Bedside ultrasonography by emergency physicians for anterior talofibular ligament injury. J Emerg Trauma Shock 2013; 6:195–198.

38.

Hua

Yang

Chen

, et al. Ultrasound examination for the diagnosis of chronic anterior talofibular ligament injury. Acta Radiol 2012; 53:1142–1145.

39.

Jones

Colaco

Fischer

, et al. Accuracy of point-of-care ultrasonography for pediatric ankle sprain injuries. Pediatr Emerg Care 2018; 34:842–847.

40.

Lee SH, Yun SJ. The feasibility of point-of-care ankle ultrasound examination in patients with recurrent ankle sprain and chronic ankle instability: Comparison with magnetic resonance imaging. Injury 2017;48:2323–2328.

41.

McCarthy

Wilson

Coltman

TP.

Anterolateral ankle impingement: findings and diagnostic accuracy with ultrasound imaging.

Skeletal Radiol 2008; 37:209–216.

42.

Milz

Steinborn

, et al. Lateral ankle ligaments and tibiofibular syndesmosis. 13-MHz high-frequency sonography and MRI compared in 20 patients. Acta Orthop Scand 1998; 69:51–55.

43.

Oae

Takao

Uchio

, et al. Evaluation of anterior talofibular ligament injury with stress radiography, ultrasonography and MR imaging. Skeletal Radiol 2010; 39:41–47.

44.

Raatikainen

Putkonen

Puranen

Arthrography, clinical examination, and stress radiograph in the diagnosis of acute injury to the lateral ligaments of the ankle.

Am J Sports Med 1992; 20:2–6.

45.

van Dijk

Lim

Bossuyt

, et al. Physical examination is sufficient for the diagnosis of sprained ankles. J Bone Joint Surg Br 1996; 78:958–962.

46.

De Maeseneer

Marcelis

Jager

, et al. Sonography of the normal ankle: a target approach using skeletal reference points. AJR Am J Roentgenol 2009; 192:487–495.

47.

Glaser

Friedl

Welk

[The value of ultrasound in the diagnosis of capsule ligament injuries of the upper ankle joint].

Unfallchirurg 1989; 92:540–546.

48.

Howard

Noble

Marill

, et al. Bedside ultrasound maximizes patient satisfaction. J Emerg Med 2014; 46:46–53.

49.

Melniker

Leibner

McKenney

, et al, Mancuso

CA.

Randomized controlled clinical trial of point-of-care, limited ultrasonography for trauma in the emergency department: the first sonography outcomes assessment program trial.

Ann Emerg Med 2006; 48:227–235.

50.

Moore

Copel

JA.

Point-of-care ultrasonography.

N Engl J Med 2011; 364:749–757.

51.

Ultrasound Guidelines: Emergency, Point-of-Care and Clinical Ultrasound Guidelines in Medicine. Ann Emerg Med 2017;69:e27–e54.