Abstract
Early diagnosis of serious bacterial infections (SBI) is important for improving outcome of morbidity and mortality in children. A systematic review was conducted to examine if shivering had any value in diagnosing serious bacterial infection. We split our population (0–18 years old) into two categories depending on the presence of a known malignancy. The databases of Medline, Embase, Cinahl, and Web of Science were searched from inception until July 2019. The quality was assessed with the QUADAS-2 tool. Two by two tables were created, extracting the number of true positive (TP), true negative (TN), false positive (FP), and false negative (FN) regarding shivering and SBI, by 2 authors independently. Sensitivity, specificity, likelihood ratios, and their 95% confidence intervals were calculated using the MetaDATA Shiny app. In a population with known malignancy, we found a +LR of 3.47 (95% CI 2.58–4.36) for a serious bacterial infection when shivering was present, implying an increase of 25–30% possibility for a serious bacterial infection. In children without malignancy, diagnostic accuracy of shivering was poor.
Conclusion: Shivering is of limited use to diagnose serious bacterial infection in children without malignancy. Nevertheless, in children with known malignancy, it can be useful as an alarm signal.
|
What is Known:
• In the NICE guidelines for febrile illness in children, “shivering” is considered as an intermediate risk factor (“amber” sign) for a serious illness.
• A systematic literature search conducted in 2007 investigating the correlation between shivering in a febrile child and the presence of a serious bacterial infection could include only one study.
What is New:
• Based on the results of this systematic review, shivering has little diagnostic value in children without malignancy but can be useful as an alarm sign of serious bacterial infection in children with known malignancy.
• In case of absence of shivering, serious bacterial infection cannot be ruled out.
|