Andrew Beckman, MS4
The Article: Doniger SJ, Komblith A. Point-of-care ultrasound integrated into a staged diagnostic algorithm for pediatric appendicitis. Pediatric Emergency Care. 2016 Jun:14
The Idea: Misdiagnosis rates of appendicitis have been shown to be as high as 28% - 57% in children less than 12 years old, leading to a time delay in making the correct diagnosis. This time delay has been shown to be directly proportional to increased risk of appendiceal perforation and higher morbidity. One imaging modality employed to improve accuracy of diagnosis is the CT scan, with sensitivities 87-100% and specificities of 83-97%. However, this comes at the risk of exposing children to great amounts of ionizing radiation, leading to the use of radiology performed ultrasound for a radiation-free imaging modality. Unfortunately, 24/7 availability of radiology-performed ultrasound does not exist at every center with an emergency department. Point of care ultrasound is already a useful tool for a variety of different conditions. This study sought to compare the accuracy of point of care ultrasound (POCUS) with radiology performed ultrasound in the diagnosis of pediatric appendicitis, in an attempt to decrease CT scans in young children.
The Study: This was a prospective cohort study conducted at a tertiary care level 1 academic emergency department with a reported census of 13,000 pediatric visits to the ED.
Inclusion criteria: Ages 2-18, presenting with abdominal pain concerning for acute appendicitis.
Exclusion criteria: Patients outside of the defined age range, pregnancy, referred to the ED with prior imaging obtained, unable to tolerate examination, and those unable/unwilling to provide consent.
Three physicians, an EM resident, EM attending (US fellowship trained) and pediatric EM attending (US fellowship trained), received a 30 minute tutorial and performed 40 quality assurance scans prior to the study. Patients were recruited by convenience sampling when any of three physicians were on shift. Forty patients were enrolled upon meeting inclusion criteria, and received POCUS followed by radiology performed ultrasound and a CT scan if radiology ultrasound was equivocal. A positive POCUS was a blind-end, non-compressible tubular structure with a diameter of greater than 6 mm and without peristalsis. A negative POCUS was a compressible appendix measuring less than 6 mm in diameter, visualized in its entirety. Other results were equivocal and required CT scan. Positive results were referred to surgery and negative results were discharged with 2 week follow up calls.
Primary Endpoint: accuracy of POCUS compared to radiology-performed ultrasound
Secondary Endpoints: effect of patient BMI and age on POCUS accuracy and number of CTs avoided.
Results: Of the 40 enrolled patients, 16 had acute appendicitis. POCUS correctly identified 15 of 16 cases, 1 scan was equivocal and needed a CT scan. Radiology ultrasound correctly identified 13 of 16 cases. POCUS correctly identified the 3 cases missed by radiology. 18 patients received a CT scan. POCUS had 3 false positives. The specificity and sensitivity of POCUS was calculated to be 93.8 % and 87.5% respectively. Patients were stratified into 2 groups based on time of enrollment and POCUS sensitivity and specificity were calculated as 87.5% and 75% respectively in patients enrolled in the first half of the study, and 100% and 100% in the second half of the study. All 3 false negative POCUS scans were done in the first half of the study. Sensitivity and specificity of radiology ultrasound was calculated as 81.25% and 100% respectively. There was good agreement between POCUS and radiology ultrasound with a κ of 0.83 (P<0.0005).
The study found that age and BMI had no effect on the accuracy of POCUS. Since radiology ultrasound superseded POCUS, it was calculated that radiology ultrasound reduced CT scans by 55%. A potential reduction in CT scans was calculated as 83% had the clinician relied on POCUS alone.
Take-away: The results of this study suggest that point of care ultrasound is comparable in accuracy to radiology-performed ultrasound in the diagnosis of pediatric appendicitis. However, certain limitations to this study exist. A major limitation is the small sample size, which affects not only the primary end-point but the secondary endpoints as well. Patients were also enrolled by convenience sampling, when at least 1 of 3 physicians trained for the study were present. This meant that not all eligible patients in the time period of the study were enrolled and that patients with a higher clinical suspicion of appendicitis were likely preferentially enrolled, creating a selection bias. Furthermore, 37 of 40 enrolled patients were enrolled by the same physician participating in the study, an issue considering that POCUS is highly user-dependent. A larger scale study is needed to further investigate the accuracy of POCUS in the diagnosis of pediatric appendicitis.