• 
• 

Three Part Question

In [pediatric patients with head injury following blunt trauma presenting to the emergency department] what is the [sensitivity and specificity of skull fracture detection] using [point of care ultrasound]?

Clinical Scenario

A two-year-old healthy male presents to the emergency department with his parents after a witnessed fall from a chair. Parents say he was initially irritable but mostly himself. An age-appropriate neurologic exam is reassuring but you notice he has a temporal hematoma. You are deciding whether to complete a head computed tomography (CT) in this child to rule out skull fracture and traumatic brain injury by reviewing decision making algorithms, such as the Pediatric Emergency Care Applied Research Network (PECARN) head injury algorithm. CT is unavailable at your hospital at this hour and the patient would need to be transported to a different center. While trying to decide, you wonder how sensitive and specific point of care ultrasound (POCUS) is in detecting skull fractures in children

Search Strategy

Ovid MEDLINE (R) and Epub Ahead of Print, In-Process, In-Data-Review & Other Non-indexed Citations and Daily from 1946 to January 26th, 2024.
[exp Skull Fractures OR exp Brain Injures OR exp Craniocerebral Trauma OR exp Fractures, Bone OR skull fractur$.mp OR head traum$.mp OR head injur$.mp] AND [exp Ultrasonography OR ultrasound.mp OR point of care ultrasound.mp] AND [exp Emergency Medicine OR exp Emergency Medical Services OR emergency room.mp OR emergency department.mp OR ER.mp OR ED.mp OR A&E.mp]. LIMIT to “all child (0 up to and less than 18 years of age)”.

Inclusion criteria: studies with pediatric patients up to and less than 18 years old with blunt head injury. For determining accurate sensitivity and specificity of skull fracture detection with POCUS, articles must report these values as compared to the current standard of diagnosis which is CT.

Exclusion criteria: studies in which POCUS was completed by non-emergency medicine professionals and patients >18 years old.

Search Outcome

A total of 162 publications were identified and screened by title and abstract. After screening and removal of duplicates, 13 studies underwent full text review. Five articles met inclusion criteria. We reviewed the bibliographies of meta-analyses and systematic reviews, identifying one additional publication for a total of six included papers.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Dehbozorgi et al. 2021 Iran	168 patients with a closed head injury requiring a head CT scan were enrolled in the study and received both a CT and skull POCUS. Recruitment age 0-14. Mean age ± standard deviation of 6.21 ± 3.99 years. Single-center study completed at a large urban academic center.	A prospective cross-sectional study where POCUS was completed by four first year emergency medicine residents, who had completed a single day training session, under the supervision of an attending physician. Scans were completed in two plains over the suspected impact site based on physical exam findings.	Sensitivity and specificity as compared to CT.	There were 11/168 patients (6.5%) with fractures identified in the study population. POCUS had a sensitivity of 81.8% (95% CI 48.2–97.7%) and a specificity of 100% (95% CI, 97.7–100%) in skull fracture detection as compared to CT.	Imprecision in sensitivity as evidenced by wide confidence interval. The decision to obtain a head CT was not standardized with a decision-making algorithm. Mechanism of injury not listed.
Choi et al. 2020 South Korea	87 patients with a closed head injury requiring a head CT scan were enrolled in the study and received both a CT and skull POCUS. Recruitment age 0-4. Mean age ± standard deviation of 21.3 ± 12.5 months. Single-center study completed at an urban, tertiary-care, level I trauma, academic university hospital.	A prospective observational study in which POCUS was completed by three emergency medicine fellows with previous experience in POCUS who also completed a one-hour training session and practice scans under supervision in preparation for this study. Scans were completed in two plains over the suspected impact site and surrounding area based on physical exam findings.	Sensitivity and specificity as compared to CT.	There were 13/87 patients (14.9%) with fractures identified in the study population. POCUS had a sensitivity of 76.9% (95% CI 46.0–93.8%) and specificity of 100% (95% CI, 93.9–100%) in skull fracture detection as compared to CT.	Imprecision in sensitivity as evidenced by wide confidence interval. The decision to obtain a head CT was not standardized with a decision-making algorithm. Small sample size.
Masaeli et al. 2019 Iran	538 patients with a closed head injury within 24 hours requiring a head CT scan (using the PECARN Head Injury algorithm or gestalt) were enrolled and received both a CT scan and skull POCUS. Recruitment age 0-18. Mean age ± standard deviation of 5.6 ± 4.9 years. Study completed at two large tertiary care hospitals.	A multi-center prospective cross-sectional study in which POCUS scans were completed by third-year emergency medicine residents (with didactic and hands-on training under supervision) and emergency medicine attending physicians. Scans were completed to assess for skull fractures and brain hemorrhage using a bilateral trans-temporal approach.	Sensitivity and specificity in diagnosing brain hemorrhages and skull fractures as compared to CT.	There were 78/538 patients (14.5%) with fractures identified in the study population. POCUS had a sensitivity of 92.3% (95% CI 84.0-97.1%) and specificity of 95.8% (95% CI 93.6-97.5%) in skull fracture detection as compared to CT.	Does not detail specific training in skull POCUS in preparation for this study. Physicians performing scans had various levels of experience with POCUS. Does not list the number of physicians completing the skull POCUS.
Parri et al. 2017 Italy and USA	115 patients with a closed head injury requiring a head CT scan (using the PECARN Head Injury algorithm or gestalt) were enrolled and received both a CT and skull POCUS. Recruitment age 0-2. Mean age ± standard deviation of 7.9 ± 6.2 months. Patients were enrolled in one general and five pediatric emergency departments.	A multi-center prospective observational study in which POCUS scans were completed by residents (with supervision) and attendings in emergency medicine and pediatrics with varying experience in POCUS. All completed two didactic video sessions and had to demonstrate competency in 10 skull POCUS examinations in children under two years old in preparation for this study. Scans were completed over the suspected impact site and surrounding area based on physical exam findings.	Sensitivity and specificity as compared to CT.	There were 88/115 patients (76.5%) with fractures identified in the study population. POCUS had a sensitivity of 90.9% (95% CI 82.9-96.0%) and specificity of 85.2% (95% CI 66.3-95.8%) in skull fracture detection as compared to CT.	Does not list the number of physicians completing the skull POCUS. Physicians performing scans had various levels of experience with POCUS. It is unclear which year of training residents were that performed skull POCUS in this study. Mechanism of injury not listed.
Parri et al. 2013 Italy	55 patients with a closed head injury requiring a head CT scan were enrolled and received both a CT and skull POCUS. Recruitment age 0-18. Mean age approximately 5 years. Single-center study completed at a pediatric emergency department	A prospective observational study in POCUS scans were completed by an emergency ultrasound fellow experienced in POCUS and six pediatric emergency medicine physicians who previously had no experience in POCUS. The six physicians inexperienced in POCUS completed a 16-hour ultrasound training course prior to this study. All scanners received additional cranial ultrasound training on volunteers in preparation for this study. Scans were completed in two plains over the suspected impact site based on physical exam findings.	Sensitivity and specificity as compared to CT.	There were 35/55 patients (63.6%) with fractures identified in the study population. POCUS had a sensitivity of 100% (95% CI 88.2–100%) and specificity of 95% (95% CI 75.0–99.9%) in skull fracture detection as compared to CT.	Physicians performing scans had various levels of experience with POCUS. Unclear how extensive was training preparation. The decision to obtain a head CT was not standardized with a decision-making algorithm.
Riera & Chen 2012 USA	46 patients with a closed head injury requiring a head CT scan were enrolled in the study and received both a CT and skull POCUS. Recruitment age 0-18. Median age of 2 years. Single-center study completed in an urban, tertiary-care, level I trauma pediatric emergency department	A prospective observational study in which POCUS scans were completed by four pediatric emergency medicine physicians with at least one month training in POCUS. Scans were completed in two plains over the suspected impact site based on physical exam findings.	Sensitivity and specificity as compared to CT.	There were 11/46 patients (23.9%) with fractures identified in the study population. POCUS had a sensitivity of 82% (95% CI 48 - 97%) and specificity of 94% (95% CI 79 - 99%) in skull fracture detection as compared to CT.	Physicians completing the skull POCUS were not always blinded to the results of the head CT if skull POCUS was completed after CT scanning introducing potential misclassification of outcome bias Limited number of physicians were able to enroll patients to this potentially contributing to small sample size. Mechanism of injury not listed. Variability in training by scanners (one had 10 years of extensive experience in pediatric ultrasound and the other three scanners had one month adult ultrasound training). No specific training in skull POCUS was completed in preparation for this study. The decision to obtain a head CT was not standardized with a decision-making algorithm. Imprecision in sensitivity as evidenced by wide confidence interval.

Comment(s)

All studies limited inclusion to patients who were receiving a head CT. This population is inherently at higher risk and may limit generalizability of these findings to lower-risk pediatric head injury patients who may not require a head CT. There was a substantial difference in fracture prevalence (ranging from 7% to 77%) across studies, suggesting a difference in the study populations. Fractures were often missed due to anatomical difficulties of the suspected fracture site (skull base and eye orbits) and when the hematoma was not aligned with the fracture site. Poor patient tolerance of scans was noted among several papers as an exclusion criterion or challenge. Differences in training and operator experience were observed among many of the selected studies. If the CT was completed first, physicians performing the skull POCUS were blinded to the results of the CT in all studies except one (Riera & Chen, 2012).

Clinical Bottom Line

In pediatric patients with a minor head injury, POCUS performed by EM physicians has a sensitivity ranging between 77-100% and a specificity between 85-100% for skull fracture detection, and its use in clinical decision-making has yet to be validated.

References

1. Dehbozorgi A, Mousavi-Roknabadi RS, Hosseini-Marvast SR, Sharifi M, Sadegh R, Farahmand F, Damghani F. Diagnosing skull fracture in children with closed head injury using point-of-care ultrasound vs. computed tomography scan. Eur J Pediatr. 2021 Feb;180(2):477-484.
2. Choi JY, Lim YS, Jang JH, Park WB, Hyun SY, Cho JS. Accuracy of Bedside Ultrasound for the Diagnosis of Skull Fractures in Children Aged 0 to 4 Years. Pediatr Emerg Care. 2020 May;36(5):e268-e273.
3. Masaeli M, Chahardoli M, Azizi S, Shekarchi B, Sabzghabaei F, Shekar Riz Fomani N, Azarmnia M, Abedi M. Point of Care Ultrasound in Detection of Brain Hemorrhage and Skull Fracture Following Pediatric Head Trauma; a Diagnostic Accuracy Study. Arch Acad Emerg Med. 2019 Sep 24;7(1):e53.
4. Parri N, Crosby BJ, Mills L, Soucy Z, Musolino AM, Da Dalt L, Cirilli A, Grisotto L, Kuppermann N. Point-of-Care Ultrasound for the Diagnosis of Skull Fractures in Children Younger Than Two Years of Age. J Pediatr. 2018 May;196:230-236.e2.
5. Parri N, Crosby BJ, Glass C, Mannelli F, Sforzi I, Schiavone R, Ban KM. Ability of emergency ultrasonography to detect pediatric skull fractures: a prospective, observational study. J Emerg Med. 2013 Jan;44(1):135-41.
6. Riera A, Chen L. Ultrasound evaluation of skull fractures in children: a feasibility study. Pediatr Emerg Care. 2012 May;28(5):420-5.