Objectives: The aim of this study was to identify risk factors for recurrent intussusception after a successful reduction by fluoroscopy-guided air enema, the time required for recurrence, and the association between delayed reduction and the recurrence. Methods: Medical records of 479 consecutive children with intussusception who underwent fluoroscopy-guided air enema between January 2004 and September 2014 were reviewed. Recurrent intussusception was defined as a recurrence within 48 hours of a reduction. Symptom-to-door time was defined as the time from symptom onset to emergency department arrival. Door-to-reduction time was defined as the time from emergency department arrival to reduction. Time-to-recurrence was defined as the time required for recurrence from the first ultrasound diagnosis. Results: Of the 360 eligible children, 32 had recurrent intussusceptions (8.9%). Multivariable logistic regression showed that age 2 years or older is an independent predictor of recurrent intussusception (odds ratio, 2.39; 95% confidence interval, 1.13–5.02; P = 0.02). Median time to recurrence was 25 hours (18.0–36.0 hours). Although symptom-to-door and door-to-reduction times tended to be longer in the recurrence group, these differences were not significant (12.5 vs 7.0 hours, P = 0.18; 154.0 vs 143.0 minutes, P = 0.67, respectively). Conclusions: Our data suggest that provision for early recurrence and extended observation may be beneficial for children 2 years or older. Delayed reduction was not associated with recurrent intussusception, but further studies with larger sample sizes are needed to explain this issue.

Objectives: The aim of this study was to evaluate whether point-of-care ultrasound (POCUS) for intussusception screening streamlines the workflow of clinically nonspecific intussusception (CNI), an intussusception presenting with only 1 manifestation of the classic triad, and/or vomiting. Methods: We reviewed 274 consecutive children with intussusception, aged 6 years or younger, who visited a tertiary care hospital emergency department between May 2012 and April 2016. This period was dichotomized by May 2014 (the “PRE” and “POST” groups), starting point of implementation of the POCUS protocol for intussusception screening. All children with CNI who had positive results on or forwent POCUS underwent radiologist-performed ultrasound (US). We measured and compared emergency department length of stay (EDLOS), the sum of door-to-reduction and observation times, and the frequency of POCUS and positive US results between the 2 groups. Results: Of 160 children with CNI, 93 visited the emergency department since May 2014. The POST group showed a shorter median EDLOS (856 vs 630 minutes, P < 0.001), door-to-reduction time (137 vs 111 minutes, P = 0.002), and observation time (700 vs 532 minutes, P < 0.001). The POST group had a higher frequency of POCUS (12% vs 60%, P < 0.001) with positive US results (33% vs 59%, P < 0.001). The PRE group had a higher frequency of severe bowel edema (16% vs 1%, P < 0.001). No significant differences were found in the severity, recurrence, admission, and surgery. One child had a false-negative result on POCUS. Conclusions: Point-of-care ultrasound could streamline the workflow of CNI via decrease in EDLOS and unnecessary referrals for US.

A previously healthy 12-month-old girl presented to the emergency department with vomiting of water beads (superabsorbent polymer). The girl did not have clinical or radiographic signs of residual foreign bodies or intestinal obstruction. Point-of-care ultrasound showed well-demarcated, round, and hypoechoic materials in the stomach and first part of the duodenum, indicating ingested beads. Subsequently, the beads were retrieved by the esophagogastroduodenoscopy. Because water beads can be readily found with point-of-care ultrasound, the use of this imaging modality can expedite endoscopic intervention and avoid surgical removal of foreign bodies.

Background: To compare admission lactate and base deficit (BD), which physiologically reflect early hemorrhagic shock, as outcome predictors of pediatric trauma. Methods: We reviewed the data of children with trauma who visited a Korean academic hospital from 2010 through 2018. Admission lactate and BD were compared between children with and without primary outcomes. The outcomes included in-hospital mortality, early (≤24 h) transfusion, and early surgical interventions for the torso or major vessels. Subsequently, performances of lactate and BD in predicting the outcomes were compared using receiver operating characteristic curves. Logistic regressions were conducted to identify the independent associations of the two markers with each outcome. Results: Of the 545 enrolled children, the mortality, transfusion, and surgical interventions occurred in 7.0%, 43.5%, and 14.9%, respectively. Cutoffs of lactate and BD for each outcome were as follows: mortality, 5.1 and 6.7 mmol/L; transfusion, 3.2 and 4.9 mmol/L; and surgical interventions, 2.9 and 5.2 mmol/L, respectively. No significant differences were found in the areas under the curve for each outcome. Of the two markers, a lactate of >5.1 mmol/L was associated with mortality (adjusted odds ratio, 6.43; 95% confidence interval, 2.61–15.84). A lactate of >3.2 mmol/L (2.82; 1.65–4.83) and a BD of >4.9 mmol/L (2.32; 1.32–4.10) were associated with transfusion, while only a BD of >5.2 mmol/L (2.17; 1.26–3.75) was done with surgical interventions. Conclusions: In pediatric trauma, lactate is more strongly associated with mortality. In contrast, BD may have a marginally stronger association with the need for hemorrhage-related procedures.

Background: To investigate the clinical features of coinfection due to Mycoplasma pneumoniae(M. pneumoniae), a common copathogen in influenza, in influenza pneumonia patients. Methods: We reviewed 4,465 patients with influenza who visited a tertiary care hospital emergency department in Seoul (Korea) from 2010 through 2016, and underwent immunoglobulin M (IgM) serology or polymerase chain reaction (PCR) for M. pneumoniae. Influenza pneumonia was defined as laboratory-confirmed influenza plus radiographic pneumonia. Patients with healthcare-associated pneumonia or non-mycoplasma bacterial coinfection were excluded. Clinical, laboratory, and radiographic findings and outcomes of the influenza pneumonia patients with and without M. pneumoniae coinfection were compared. Multivariable logistic regression analysis was performed to identify factors associated with the coinfection. Results: Of 244 influenza pneumonia patients, 41 (16.8%) had M. pneumoniae coinfection. These patients were younger with a higher frequency of age of 5–10 years, and had higher white blood cell (WBC) and lymphocyte counts; lower concentration of C-reactive protein (CRP). The coinfection had no association with specific radiographic findings and poor outcome. Multivariable analysis showed the age of 5–10 years (adjusted odds ratio, 18.83; 95% confidence interval, 5.899–60.08; P<0.001) as the factor associated with the coinfection. Conclusions: M. pneumoniae coinfection in influenza pneumonia may be associated with the age of 5–10 years, and otherwise clinically indistinct from influenza pneumonia without the coinfection. This finding suggests the need for M. pneumoniae testing in patients aged 5–10 years with influenza pneumonia.

Objectives: The aim of the study was to investigate the diagnostic efficacy of C-reactive protein (CRP) in predicting serious bacterial infection (SBI) in febrile children aged 3 to 36 months with extreme leukocytosis (EL), defined as a peripheral white blood cell count of 25,000 to 49,999/mm³. Methods: The presence of SBI was reviewed in previously healthy children aged 3 to 36 months with a fever of 39°C or higher and EL, who visited a tertiary care hospital emergency department between September 2010 and August 2015. We measured cutoff values of CRP with corresponding likelihood ratios (LRs) and posttest probabilities (PPs). Results: Of 9989 febrile children, 4252 (42.6%) underwent assays for white blood cell and CRP. Of 233 (5.5%) children with EL, 133 (3.1%) were enrolled, of whom 43 (32.3%; 95% confidence interval [CI], 25.0%–40.7%) had SBI, including 33 pyelonephritis, 5 deep abscesses, 3 lobar pneumonia, and 2 soft tissue infections. With the cutoff of 7.8 mg/dL, the sensitivity of CRP in predicting SBI was 81.4% (95% CI, 67.4%–90.3%); the specificity, 80.0% (95% CI, 70.6%–87.0%); the positive predictive value, 66.0% (95% CI, 52.6%–77.3%); the negative predictive value, 90.0% (95% CI, 81.5%–94.9%); the positive LR, 4.1; and the PP, 66.0% (95% CI, 55.6%–75.0%). A CRP concentration of 16.1 mg/dL or higher yielded an LR of 11.2 and a PP of 84.2% (95% CI, 62.1%–94.5%). A CRP concentration of lower than 3.4 mg/dL yielded an LR of 0.05 and a PP of 2.4% (95% CI, 0.3%–14.6%). Conclusions: High concentration of CRP is strongly associated with the presence of SBI in febrile children with EL.