Point‐of‐Care Ultrasound for the Diagnosis of Hip Effusion in Children Presenting to the Emergency Department With Atraumatic Limp or Hip Pain: A Scoping Review

Abstract

This study reviews the evidence for point‐of‐care ultrasound (POCUS) for the detection of hip effusion in children presenting to the emergency department with atraumatic limp or hip pain. A scoping review was conducted using the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses Extension for Scoping Review (PRISMA‐ScR) guideline. PubMed, EMBASE, and the Cochrane Library databases were searched from January 2000 to October 2025. All study designs were considered. The primary outcome was diagnostic accuracy of POCUS for hip effusion, and key secondary outcomes included provider characteristics and clinical integration. Data were also charted across study, participant, and ultrasound characteristics. Thirteen studies met inclusion criteria: 3 prospective studies, 9 retrospective cohorts, and 1 case report. Studies predominately evaluated diagnostic accuracy of POCUS performed by emergency medicine physicians with variable training compared with radiology performed ultrasound, feasibility of POCUS guided hip arthrocentesis, and explored its integration into clinical decision rules. Diagnostic accuracy was high, with reported sensitivity ranging from 80% to 98% and specificity from 91% to 98%. This scoping review demonstrates that there is growing evidence that hip POCUS is a useful clinical tool for evaluation of pediatric patients presenting to the emergency department with atraumatic limp or hip pain. It has been shown to be accurate when performed by emergency physicians after variable training. Further research is required with large prospective multicenter studies to explore patient‐centered outcomes and validate diagnostic pathways that incorporate POCUS.

Abbreviations

ED

emergency department

PEM

pediatric emergency medicine

POCUS

point‐of‐care ultrasound

Atraumatic limp or hip pain in children are common presentations to the emergency department (ED). ¹ Pediatric patients with limp pose a diagnostic challenge owing to the unique differences in how symptoms may present, including non‐specific complaints and difficulty localizing pain. Younger children can also be difficult to examine, with cooperation often hindered due to pain, anxiety, or limited understanding. ² The cause of pediatric atraumatic limp or hip pain can range from self‐limiting conditions such as transient synovitis to more serious conditions such as septic arthritis, osteomyelitis, or malignancy. ³

One means of narrowing down the differential diagnosis is by diagnosis of hip effusion using ultrasound. ⁴ Hip effusions are predominantly found in patients with transient synovitis or septic arthritis but unfortunately cannot reliably discriminate between these pathologies. ⁵ Ultrasound is more sensitive than X‐ray for the diagnosis of hip effusions and more accessible than MRI, with the added advantage of being inexpensive and conducted at the bedside without the requirement for sedation. ⁶ Point‐of‐care ultrasound (POCUS) is readily available in many EDs as an adjunct to aid clinical decision making by allowing physicians to integrate clinical examination findings with real time sonographic imaging. ⁷ The objective of this scoping review was to describe the evidence for POCUS for the detection of hip effusion in children presenting to the ED with atraumatic limp or hip pain.

Methods

Study Design

A scoping review was conducted and reported using the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses Extension for Scoping Review (PRISMA‐ScR) checklist ⁸ (online supplemental Appendix 1).

Search Strategy

A comprehensive search was undertaken in conjunction with a hospital librarian who assisted with designing the search strategies. PubMed, EMBASE, and the Cochrane Library databases were searched on September 1, 2025 and October 30, 2025 using predefined variables (online supplemental Appendix 2). Publications were searched from January 1, 2000 to October 30, 2025 and only studies in the English language were included. All study designs were eligible for inclusion, including randomized controlled trials, cohort studies, case‐control studies, cross‐sectional studies, case series, and case reports.

Study Selection

Search results were imported into Refworks. Two reviewers (Nigel Chan and David J. McCreary) independently identified potentially eligible studies for inclusion by manual screening by title and abstract, removal of duplicates, and then full‐text review of the remaining studies. A third reviewer (Peter J. Snelling) was available for consultation in the event of any disagreements that could not be resolved by consensus. Publications considered eligible for this review were any that reported the use of POCUS for children, or if they included a subset aged <18 years, with separate data.

Outcome Measures

The primary outcome measure was diagnostic accuracy of POCUS for hip effusion in children presenting with atraumatic limp or hip pain, compared with an appropriate reference standard. Secondary outcomes included ultrasound‐related outcomes such as participant characteristics, providers and training received, ultrasound characteristics, and integration into diagnostic or procedural pathways.

Data Extraction

A standardized form was used for data extraction of study characteristics and outcomes. Two study authors (Nigel Chan and David J. McCreary) independently extracted the data, with any unresolved disagreements referred to a third reviewer (Peter J. Snelling) for resolution. Data extracted included study characteristics (design, location, setting, sample size), participant characteristics (demographics, presentation, diagnosis), ultrasound characteristics (training, operators, equipment), and outcomes of interest (diagnostic accuracy, clinical integration).

Data Analysis

Quantitative data were described using summary statistics. Categorical variables were summarized as frequency (percentage) and continuous variables were summarized as either mean (standard deviation) or median (interquartile range) depending on distribution.

Results

Study Selection and Characteristics

The database searches yielded a total of 197 articles, with 53 duplicates removed and a further 118 articles excluded after title and abstract screening. Full‐text evaluation of the remaining 26 publications was undertaken with 13 studies fulfilling inclusion criteria. One additional study was added during an audit of articles within the time period that was searched. ⁹ One of the 14 studies was removed as it included data on adults and children and it was not possible to separate the pediatric data. ¹⁰ The study authors were contacted to obtain pediatric‐specific data but no response was obtained, resulting in 13 studies included in the scoping review (Figure 1).

Figure 1.

Preferred Reporting Items for Systematic reviews and Meta‐Analyses conducted on September 1, 2025 and October 30, 2025.

A total of 2062 participants across 13 eligible studies were included. Table 1 summarizes the study characteristics. Studies were mostly conducted in the United States (6) and Israel (5), Australia (2), one of which was a multicenter study with sites in both Australia and the United States, and a single study conducted in the United Kingdom. Study designs included 3 prospective studies (1 multicenter), 9 retrospective cohorts, and 1 case report. All involved EDs in either tertiary hospitals (pediatric or mixed) or large district general hospitals (mixed).

Table 1.

Characteristics of Included Studies

Reference (Primary Author)	Publication Year	Recruitment Period	Country	Study Design	Centers	Department	Setting
Jones	2025	October 2019–October 2023	USA, Australia	Prospective diagnostic accuracy	5 hospital sites	Pediatric emergency departments	Tertiary pediatric hospital and Large mixed hospital with dedicated pediatric emergency department
McCreary	2025	May 2019–May 2024	UK	Retrospective cohort	Single	Pediatric emergency department	Large district general hospital
Katz‐Dana	2024	October 1, 2020–September 15, 2021	Israel	Prospective observational	Single	Pediatric emergency department	Tertiary pediatric hospital
Bailey	2024	May 2016–May 2021	Australia	Retrospective cohort	Single	Pediatric emergency department	Tertiary pediatric hospital
Berant	2022	2014–2019	Israel	Retrospective case series	Single	Pediatric emergency department	Tertiary pediatric hospital
Zoabi	2021	January 1, 2014–December 31, 2019	Israel	Retrospective case series	Single	Pediatric emergency department	Tertiary pediatric hospital
Gravel	2021	October 2015–May 2018	USA	Retrospective cohort	Single	Pediatric emergency department	Tertiary pediatric hospital
Cruz	2018	2009–2014	USA	Retrospective cohort	Single	Pediatric emergency department	Tertiary pediatric hospital
Garrison	2016	2016	USA	Case report	Single	Pediatric emergency department	Tertiary pediatric hospital
Kotlarsky	2016	January 1, 2007–December 31, 2014	Israel	Retrospective case series	Single	Emergency department	Tertiary pediatric hospital
Viera	2010	April 2007–December 2007	USA	Prospective diagnostic accuracy	Single	Emergency department	Tertiary pediatric hospital
Tsung	2007	Not specified	USA	Case series	Single	Emergency department	Tertiary pediatric hospital
Shavit	2006	Not specified	Israel	Case series	Single	Emergency department	Tertiary pediatric hospital

Characteristics of Study Participants

Characteristics of study participants are summarized in Table 2. The mean age of children ranged from 5.5 to 6.4 years. The range of ages was 0–18 years, with the exception of one study which included 6 (1%) participants aged up to 21 years, which is reflective of the practice in their department. ¹¹ There was a predominance in the male sex (62%). Sample sizes ranged from 1 to 621. Presenting features included atraumatic limp, hip pain, and refusal to weight bear. The etiologies of hip effusion were transient synovitis, septic arthritis, osteomyelitis, Perthes disease, Lyme disease, and spondyloarthropathy.

Table 2.

Characteristics of Study Participants

Reference (Primary Author)	Sample Size (Recruited)	Age (Years)	Sex	Participant Characteristics	Etiology of Hip Effusions
Jones	161	0–15 (mean 6.4)	65.8% M	Atraumatic limp, hip pain, or refusal to bear weight	77 hip effusions: 2 septic arthritis, 6 Lyme disease, 69 transient synovitis
McCreary	178	1–10	58.4% M	Atraumatic limp	178 patients primarily coded as transient synovitis
Katz‐Dana	95	Mean 6.4	70.5% M	New limp, lower limb pain, or refusal to bear weight	49 hip effusions: transient synovitis, septic arthritis, other (numbers not specified)
Bailey	190	0–18	Not specified	Atraumatic limp or “suspicion of hip pathology”	35 septic arthritis
Berant	10	0–18 (mean 4.2)	40% M	Diagnose of septic arthritis and underwent aspiration of the hip joint by an emergency physician	Septic arthritis
Zoabi	621	3–10 (mean 5.5)	69% M	Nontraumatic hip tenderness	6 septic arthritis, 4 osteomyelitis, 4 Perthes disease, 539 transient synovitis, 3 other
Gravel	238	0–18 (median 6.5)	60% M	Atraumatic hip pain, limp or refusal to weight bear	Not specified
Cruz	516	0–22 (mean 5.8)	65% M	Atraumatic limp, leg pain, hip pain, refusal to bear weight, or had findings of hip pathology on physical examination	Not specified
Garrison	1	11	F	Atraumatic limp, hip pain and fever	Septic arthritis
Kotlarsky	14	0–17 (median 2.5)	50% M	Diagnosis of septic arthritis based on analysis of fluid aspirated from the hip joint in the ED	Septic arthritis
Viera	28	2–16 (median 7.5)	64% M	Not specified	Not specified
Tsung	5	5–11	80% M	Atraumatic hip pain	3 transient synovitis, 1 septic arthritis, 1 spondyloarthropathy
Shavit	5	2–10	Not specified	Atraumatic acute limp (less than 3 days)	3 transient synovitis, 1 septic arthritis, 1 osteomyelitis

Ultrasound Characteristics

Table 3 outlines the ultrasound characteristics. Most studies involved pediatric emergency medicine (PEM) physicians as the operator ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ with the remainder being emergency physicians. ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ Training was reported in 7 studies and varied considerably, from 30 minutes through to a full day of didactic teaching, with logbooks of 10–25 scans. ¹² , ¹³ , ¹⁴ , ¹⁵ Training for one study occurred as a 3‐month apprenticeship with an orthopedic surgeon. ¹⁶ High frequency linear transducers were used in all studies, with a minimum frequency of 5 MHz. Measurement criteria were similar across studies, with the definition of the measurement between the anterior surface of the femoral neck and the posterior aspect of the iliopsoas muscle (>5 mm or more than 2 mm asymmetry compared with the opposite asymptomatic hip). ⁹ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ However, the placement of callipers when measuring the effusion varied, with some ¹¹ , ¹³ , ¹⁹ measuring the internal anechoic stripe (ie, between the thickened anterior and posterior synovium), one ²⁰ measured from the top of internal anechoic stripe to the bone (ie, included the posterior synovium but excluded the anterior synovium), and others measured the entire hip capsule (ie, posterior margin of the psoas to the bone) ¹⁴ , ¹⁵ , ¹⁶ in the figures provided. Diagnostic accuracy ranged from a sensitivity of 80.0%–98.1% and a specificity of 91.3%–98.0%. ¹¹ , ¹² , ¹⁴ , ¹⁵ Misclassifications were mainly attributed to suboptimal imaging or capsule measurement errors. One study reported the overall accuracy as a function of POCUS experience increased by 2.4% per scan. ¹¹

Table 3.

Ultrasound Characteristics

Reference (Primary Author)	Operators	Training	Transducer Frequency	Reference Standard	Performance Criteria
Jones	18 PEM physicians	Credentialing described as physicians who had completed ultrasound training as part of their pediatric emergency fellowship, an ultrasound fellowship, an accredited sonography training program, or an hour‐long didactic and hands‐on session and a logbook of 25 scans	5–10 MHz or 6–15 MHz	RADUS	Sensitivity 89.3% (95% CI 80.6–95.0) Among 3 high‐volume physicians (61.5% of all scans): 98.1% (95% CI 88.8–99.9) Remaining 15 physicians: 83.3% (95% CI 61.8–94.5) Specificity 99.2% (97.0–99.9%) Among 3 high‐volume physicians (61.5% of all scans): 97.8% (95% CI 86.8–99.9). Remaining 15 physicians: PPV 97.4% (95% CI 90.4–99.3) NPV 96.3% (95% CI 93.4–98.0)
McCreary	PEM physicians	Full‐day face‐to‐face pediatric POCUS course and deemed competent following 15 supervised pediatric hip scans	12 MHz	Not applicable	Accuracy of clinical decision rule for diagnosis of transient synovitis Correctly applied for 63/67 patients (94%)
Katz‐Dana	8 PEM physicians	2‐hour didactic session, followed by 6 supervised hip scans and then logbook of 10 hip scans (including at least 1 positive scan)	Not specified	RADUS kappa = 0.81 (95% CI 0.70–0.93)	Sensitivity 89.8% (95% CI 77.7–96.6%) Specificity 91.3% (95% CI 79.2–97.5%)
Bailey	Not specified	Not specified	Not specified	Not applicable	Not applicable
Berant	1 Emergency physician	POCUS fellowship trained	5–14 MHz	Not applicable	Not applicable
Zoabi	38 Emergency physicians	Not specified	Not specified	Not applicable	Accuracy of clinical decision rule for diagnosis of transient synovitis: Sensitivity 90.9% (95% CI 88.3–93.1%) Specificity 78.6% (95% CI 60.5–89.8%) PPV 98.9% (95% CI 97.6–99.5%) NPV 28.9% (95% CI 20.0–40.0%)
Gravel	PEM physicians	Not specified	Not specified	RADUS	28 patients received both POCUS and RADUS. Overall agreement for diagnosis of effusion was 92.6% (kappa = 0.82)
Cruz	72 PEM physicians	Not specified	Not specified	PEM POCUS expert's interpretation of the scan	Sensitivity 85% (95% CI 79–89%) Specificity 98% (CI 96–99%) PPV 91% (95% CI 86–85%) NPV 93% (95% CI 90–95%)
Garrison	Emergency physician	Not specified	Not specified	Not applicable	Not applicable
Kotlarsky	Emergency physicians	Not specified	7–13 MHz	Not applicable	Not applicable
Viera	3 PEM physicians	30 min didactic session, followed by supervised scans on pediatric volunteers. Considered credentialed following logbook of 10 hip scans (with at least 1 positive hip scan)	Not specified	RADUS	Sensitivity 80% (95% CI 51–95) Specificity 98% (95% CI 85–99) PPV 92% (95% CI 60–90) NPV 88% (95% CI 60–98)
Tsung	Emergency physicians	Not specified	7.5 MHz	Not applicable	Not applicable
Shavit	2 PEM physicians	Described as spending 3 months in the orthopedic clinic performing hip ultrasonography under the instruction of an orthopedic surgeon	7.5 MHz	Not applicable	Not applicable

RADUS, radiology‐performed ultrasound; PEM, pediatric emergency medicine.

Two studies specifically evaluated the performance of a POCUS clinical decision tool to diagnose transient synovitis. ¹² , ¹⁸ Both studies involved moderately sized cohorts of participants in which a POCUS clinical decision rule was retrospectively applied to evaluate their ability to correctly classify transient synovitis in patients presenting with atraumatic limp. Both clinical decision rules performed well with few false positives in their respective cohorts, with one study reporting a sensitivity of 90.9% (95% CI 88.3–93.1). ¹⁸ Both studies hypothesized that POCUS potentially allowed clinicians to safely limit the number of blood tests in children and unnecessary radiographs, while maintaining diagnostic confidence.

Three studies evaluated the use of POCUS guided aspiration of the hip joint in suspected septic arthritis in the ED. ¹⁷ , ²⁰ , ²¹ They were small retrospective case series which concluded that this can be performed safely by physicians in the ED which may result in quicker diagnosis and initiation of treatment. Finally, one study retrospectively evaluated the economic impact of POCUS to diagnosis septic arthritis of the hip in children compared to standard care. ²² They concluded that incorporating POCUS results in expediting time to diagnosis in suspected septic arthritis of the hip and was associated with a significant cost reduction potential.

Discussion

Although limited, the available evidence demonstrates that POCUS performed by ED clinicians has high diagnostic accuracy for detecting hip effusions in children. The highest quality evidence came from a single multicenter prospective diagnostic study, where 3 PEM physicians performed almost two‐thirds of all scans. ¹⁴ Specificity was consistently high across studies, implying that POCUS is useful in ruling‐in the diagnosis of hip effusion. The sensitivity was moderate to high, with improved performance as the operator experience increased. ¹¹ , ¹² , ¹⁴ , ¹⁵ There was minimal discordance between hip POCUS performed by trained emergency physicians and radiology performed ultrasonography interpreted by sonographers. This high level of agreement supports the use of clinician performed hip ultrasound, particularly in settings with limited access to radiology services or delayed sonographer availability, provided appropriate training and governance structures are in place.

Training was varied throughout the studies where it was reported, ranging from brief 30‐minute tutorial to a 3‐month apprenticeship. Clinicians with limited prior ultrasound experience achieved competent diagnostic performance after targeted educational sessions. This is supported by a separate study where students with no prior POCUS experience were able to improve their knowledge in performing hip POCUS, identify normal sonographic anatomical landmarks and accurately measure the amount of fluid in the joint space following a brief didactic 20‐minute educational training session. ²³ Additionally, remote video training of participants with no prior POCUS experience to perform hip POCUS appears feasible. ²⁴ Further research is required as to the optimal training required for performing hip POCUS.

With high diagnostic accuracy demonstrated for hip POCUS, its incorporation into clinical decision‐making warrants consideration. POCUS may support earlier diagnosis; however, its impact on downstream resource utilization remains uncertain. Although it has been hypothesized that POCUS could reduce unnecessary investigations, alternative scenarios are plausible in which additional imaging or laboratory testing is prompted. ¹² , ¹⁸ Importantly, hip effusion alone cannot reliably distinguish transient synovitis from other causes of atraumatic hip pain. When clinically appropriate, POCUS may expedite arthrocentesis and initiation of treatment, particularly in settings where radiology or orthopedic services are unavailable or delayed. ¹⁷ , ²⁰ , ²¹ The effect of POCUS on the use of additional imaging, blood tests, and overall costs has not been adequately measured and may be paradoxical, as seen with other diagnostic interventions. These outcomes should be incorporated into future prospective studies evaluating the role of hip POCUS within structured clinical decision pathways.

This scoping review identified the need for more high‐quality prospective studies with patient‐centered outcomes and standardization of training. In particular, a multicenter randomized controlled trial is needed to evaluate the impact of hip POCUS on clinical accuracy, cost, and resource utilization, ideally compared with usual care based on clinical reasoning for patients presenting with atraumatic limp. Prospective validation of POCUS decision support algorithms across different populations and settings is also required.

Limitations

This scoping has synthesized the existing evidence on the use of POCUS for detecting pediatric hip effusions. A comprehensive search strategy was utilized in collaboration with a medical librarian and applied across major databases, ensuring broad coverage of the available evidence. Studies prior to the year 2000 were not included due to the technological advancement in ultrasound and the development of POCUS after this time but were unlikely to have missed any important studies. However, there were only a small number of studies included that were heterogeneous in design, population, operator experience, and outcomes measures, limiting the ability to directly compare findings or perform quantitative analysis. Most studies were single‐center and small in scale, reducing the generalizability of findings across diverse health systems. Patient‐centered outcomes, such as pain, patient satisfaction, and functional recovery, were not reported, which restricts conclusions on the broader impact of POCUS.

Conclusion

This scoping review provides a comprehensive synthesis of the evidence on the utility of POCUS for the diagnosis of hip effusions in children presenting to the ED with atraumatic limp or hip pain. There is growing evidence that POCUS has a high diagnostic accuracy for hip effusion when performed by emergency physicians after variable training, with particular utility as a rule‐in tool. Its integration into clinical pathways has the potential to reduce unnecessary investigations and expedite diagnosis, particularly in distinguishing transient synovitis from other causes. The current evidence is limited by a predominance of small, single‐center studies, methodological heterogeneity, and a lack of standardized training and credentialing frameworks. Importantly, patient‐centered outcomes and cost‐effectiveness analyses remain underexplored.

Supporting information

Supplemental Appendix 1 Preferred Reporting Items for Systematic reviews and Meta‐Analyses extension for Scoping Reviews (PRISMA‐ScR) Checklist.

Supplemental Appendix 2. Search strategy for Medline.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.