Which Febrile Children With Sickle Cell Disease Need a Chest X-Ray?

Machine learning risk stratification for high-risk infant follow-up of term and late preterm infants

BACKGROUND

Term and late preterm infants are not routinely referred to high-risk infant follow-up programs at neonatal intensive care unit (NICU) discharge. We aimed to identify NICU factors associated with abnormal developmental screening and develop a risk-stratification model using machine learning for high-risk infant follow-up enrollment.

METHODS

We performed a retrospective cohort study identifying abnormal developmental screening prior to 6 years of age in infants born ≥34 weeks gestation admitted to a level IV NICU. Five machine learning models using NICU predictors were developed by classification and regression tree (CART), random forest, gradient boosting TreeNet, multivariate adaptive regression splines (MARS), and regularized logistic regression analysis. Performance metrics included sensitivity, specificity, accuracy, precision, and area under the receiver operating curve (AUC).

RESULTS

Within this cohort, 87% (1183/1355) received developmental screening, and 47% had abnormal results. Common NICU predictors across all models were oral (PO) feeding, follow-up appointments, and medications prescribed at NICU discharge. Each model resulted in an AUC > 0.7, specificity >70%, and sensitivity >60%.

CONCLUSION

Stratification of developmental risk in term and late preterm infants is possible utilizing machine learning. Applying machine learning algorithms allows for targeted expansion of high-risk infant follow-up criteria.

IMPACT

This study addresses the gap in knowledge of developmental outcomes of infants ≥34 weeks gestation requiring neonatal intensive care. Machine learning methodology can be used to stratify early childhood developmental risk for these term and late preterm infants. Applying the classification and regression tree (CART) algorithm described in the study allows for targeted expansion of high-risk infant follow-up enrollment to include those term and late preterm infants who may benefit most.

Risk-score based strategy to minimize antibiotic exposure in children with sickle cell disease and fever

Severe bacterial infections (SBI) have become less frequent in children with sickle cell disease (SCD) in the last decades. However, because of their potential risk of SBI, they usually receive empirical therapy with broad-spectrum antibiotics when they develop fever and are hospitalized in many cases. We performed a prospective study including 79 SCD patients with fever [median age 4.1 (1.7–7.5) years, 78.5% males; 17 of the episodes were diagnosed with SBI and 4 of them were confirmed] and developed a risk score for the prediction of SBI. The optimal score included CRP > 3 mg/dl, IL-6 > 125 pg/ml and hypoxemia, with an AUC of 0.91 (0.83–0.96) for the prediction of confirmed SBI and 0.86 (0.77–0.93) for possible SBI. We classified the patients in 3 groups: low, intermediate and high risk of SBI. Our risk-score-based management proposal could help to safely minimize antibiotic treatments and hospital admissions in children with SCD at low risk of SBI.

What is the role of chest X-ray imaging in the acute management of children with sickle cell disease?

Children with sickle cell disease (SCD) frequently present to hospital acutely unwell and are often exposed to diagnostic chest X-rays (CXRs). Little evidence exists to determine when CXRs are clinically useful. Using electronic hospital records, we audited CXR use in children aged 0-18 who presented to hospital over the past 10 years in both an inpatient and emergency department setting. From a total of 915 first CXRs, only 28·2% of CXRs (n = 258) had clinically significant findings that altered management or final diagnosis. Of these abnormalities, consolidation represented 52·3%, effusion 8·9%, cardiomegaly 8·4% and sickle cell-related bone changes 6·3%. Indications for CXR of respiratory distress (OR = 3·74, 95% CI 2·28-6·13), hypoxia (OR = 1·86, 95% CI 1·50-2·31) and cough (OR = 1·64, 95% CI 1·33-2·02), were more likely to have significant CXR findings. Patients who had higher peak fever (38·4°C vs. 37·4°C, P = 0·001), higher peak CRP (156·4 vs. 46·1, P < 0·001) and higher WCC (20·2 vs. 13·6, P < 0·001) were more likely to have clinically significant abnormalities on CXR. We found a decision tool using either hypoxia, cough, respiratory distress, T > 38°C, CRP > 50 or WCC > 15 × 10⁹ /l as indications for CXR, to have a sensitivity of 88% (with 95% CI 0·78-0·95) and specificity of 46% (95% CI 0·43-0·50) for clinically significant findings.

Association of Guideline-Adherent Antibiotic Treatment With Readmission of Children With Sickle Cell Disease Hospitalized With Acute Chest Syndrome

IMPORTANCE

Acute chest syndrome (ACS) is a common, serious complication of sickle cell disease (SCD) and a leading cause of hospitalization and death in both children and adults with SCD. Little is known about the effectiveness of guideline-recommended antibiotic regimens for the care of children hospitalized with ACS.

OBJECTIVES

To use a large, national database to describe patterns of antibiotic use for children with SCD hospitalized for ACS and to determine whether receipt of guideline-adherent antibiotics was associated with lower readmission rates.

DESIGN, SETTING, AND PARTICIPANTS

Retrospective cohort study including 14 480 hospitalizations in 7178 children (age 0-22 years) with a discharge diagnosis of SCD and either ACS or pneumonia. Information was obtained from 41 children's hospitals submitting data to the Pediatric Health Information System from January 1, 2010, to December 31, 2016.

EXPOSURES

National Heart, Lung, and Blood Institute guideline-adherent (macrolide with parenteral cephalosporin) vs non-guideline-adherent antibiotic regimens.

MAIN OUTCOMES AND MEASURES

Acute chest syndrome-related and all-cause 7- and 30-day readmissions.

RESULTS

Of the 14 480 hospitalizations, 6562 (45.3%) were in girls; median (interquartile range) age was 9 (4-14) years. Guideline-adherent antibiotics were provided in 10 654 of 14 480 hospitalizations for ACS (73.6%). Hospitalizations were most likely to include guideline-adherent antibiotics for children aged 5 to 9 years (3230 of 4047 [79.8%]) and declined to the lowest level for children 19 to 22 years (697 of 1088 [64.1%]). Between-hospital variation in antibiotic regimens was wide, with use of guideline-adherent antibiotics ranging from 24% to 90%. Children treated with guideline-adherent antibiotics had lower 30-day ACS-related (odds ratio [OR], 0.71; 95% CI, 0.50-1.00) and all-cause (OR, 0.50; 95% CI, 0.39-0.64) readmission rates vs children who received other regimens (cephalosporin and macrolide vs neither drug class).

CONCLUSIONS AND RELEVANCE

Current approaches to antibiotic treatment in children with ACS vary widely, but guideline-adherent therapy appears to result in fewer readmissions compared with non-guideline-adherent therapy. Efforts to increase the dissemination and implementation of SCD treatment guidelines are warranted as is comparative effectiveness research to strengthen the underlying evidence base.

The prevalence of abnormal leukocyte count, and its predisposing factors, in patients with sickle cell disease in Saudi Arabia

Introduction

High white blood cell (WBC) count is an indicator of sickle cell disease (SCD) severity, however, there are limited studies on WBC counts in Saudi Arabian patients with SCD. The aim of this study was to estimate the prevalence of abnormal leukocyte count (either low or high) and identify factors associated with high WBC counts in a sample of Saudi patients with SCD.

Methods

A cross-sectional and retrospective chart review study was carried out on 290 SCD patients who were routinely treated at King Fahad Hospital in Hofuf, Saudi Arabia. An interview was conducted to assess clinical presentations, and we reviewed patient charts to collect data on blood test parameters for the previous 6 months.

Results

Almost half (131 [45.2%]) of the sample had abnormal leukocyte counts: low WBC counts 15 (5.2%) and high 116 (40%). High WBC counts were associated with shortness of breath (P=0.022), tiredness (P=0.039), swelling in hands/feet (P=0.020), and back pain (P=0.007). The mean hemoglobin was higher in patients with normal WBC counts (P=0.024), while the mean hemoglobin S was high in patients with high WBC counts (P=0.003). After adjustment for potential confounders, predictors of high WBC counts were male gender (adjusted odds ratio [aOR]=3.63) and patients with cough (aOR=2.18), low hemoglobin (aOR=0.76), and low heart rate (aOR=0.97).

Conclusion

Abnormal leukocyte count was common: approximately five in ten Saudi SCD patients assessed in this sample. Male gender, cough, low hemoglobin, and low heart rate were associated with high WBC count. Strategies targeting high WBC count could prevent disease complication and thus could be beneficial for SCD patients.

Hot Off the Press: Which Febrile Children With Sickle Cell Disease Need a Chest X-Ray?

This retrospective chart review examined the rate of acute chest syndrome (ACS) in febrile children (aged 3 months to 21 years) with sickle cell disease and used recursive partitioning to determine which clinical factors were predictive of a diagnosis of ACS. Over the course of 2 years, 697 children made 1,837 visits to one of two pediatric emergency departments. ACS was diagnosed in 185 (10%) of the visits.