ESPE Abstracts

Importance: Current practice of doubling or trebling the basal glucocorticoid (GC) replacement dose during acute illnesses in children and young people (CYP) with GC deficiency is extrapolated from research in adults and not based on robust evidence. Endogenous cortisol levels during acute illnesses in CYP without GC deficiency can provide insight for optimal GC stress dosing for CYP with GC deficiency.

Objective: To summarise published evidence for endogenous cortisol levels during a range of acute illnesses in CYP not known to have GC deficiency.

Data source: Using PRISMA guidelines, a literature search was conducted of publications between 1 January 2000 to 30 June 2020 from CINAHl, Cochrane library, Cochrane Database of Systematic Reviews, Embase and Medline.

Study selection: Two reviewers independently screened the abstracts for eligibility criteria and resolved differences by joint discussion. Inclusion criteria were: common acute illnesses; age 1 month to 18 years; basal blood cortisol levels within 48 hours of presentation. Exclusion criteria were: fewer than 5 subjects; subjects known to have GC deficiency or treatment with potential influence on cortisol levels.

Data extraction and synthesis: Two separate pairs of reviewers extracted and checked data for predefined fields. Cortisol values in traditional units were converted to SI units (nmol/l) and reported as mean and standard deviation (SD). Overall weighted means and pooled SDs were calculated.

Main outcomes and measures: Primary outcome was mean cortisol level and spread for controls and for patients in different illness groups.

Results: From 345 records screened, 15 studies were included. All were hospital based, 14 were prospective observational, one was part of a trial and five had controls. Mean cortisol levels were higher in all acute illness groups (n=689) than in controls (n=175) (811.10 vs 288.04, P<.001). Levels were highest in bacterial meningitis (mean 1281.26) and were over 3-fold higher than controls in severe gastroenteritis (mean 1094.00). Among the subgroups with sepsis, those with shock had lower cortisol levels than those without (mean 768.09 vs 1021.30, P=.002) but levels in survivors did not differ from non-survivors (mean 842.75 vs 686.84, P=.503).

Conclusions and Relevance: In children and young people, circulating cortisol levels are higher during acute illness compared to controls and vary across the range of illnesses. Further study is required to optimise GC stress dosing tailored to the nature and severity of the acute illness.