PD100 - COMPARISON OF ANTIMICROBIAL CONSUMPTION IN CRITICALLY ILL CHILDREN USING DAILY DEFINED DOSES AND DAYS OF THERAPY (ID 1063) - The 40th Annual Meeting of the European Society for Paediatric Infectious Diseases

PD100 - COMPARISON OF ANTIMICROBIAL CONSUMPTION IN CRITICALLY ILL CHILDREN USING DAILY DEFINED DOSES AND DAYS OF THERAPY (ID 1063)

AS01.d. Antibiotic Stewardship

PD100 - COMPARISON OF ANTIMICROBIAL CONSUMPTION IN CRITICALLY ILL CHILDREN USING DAILY DEFINED DOSES AND DAYS OF THERAPY (ID 1063)

Session Name

0330 - Poster Discussion Session 05: Antimicrobials & Antibiotic Stewardship (ID 6)

Presenter

Kyriaki Moustakidou Papadaki (Greece)

Authors

Abstract

Abstract

Backgrounds:

Although antimicrobial use monitoring is essential, computerized clinical records at patient level may not be available in many countries. Days of Therapy (DOTs) are recommended by WHO as an antimicrobial consumption metric but access to patient level data is required. We compared daily-defined doses (DDDs) to DOTs as antibiotic consumption metrics in critically ill children and evaluated their relation.

Methods

Data on antimicrobial consumption in a pediatric intensive care unit (PICU) from June 2018 to December 2019 were expressed as monthly DDDs per 100bed-days (DDDs/100BD) and monthly DOTs/100BD. All analyses were performed in R version4.0.4. The time series of DOTs and DDDs were examined for evidence of seasonality, stationarity and autocorrelation. A linear regression model was developed to predict DOTs/100BD using DDDs/100BD for total antimicrobial consumption and for specific antimicrobials.

Results:

During study period 178 admissions occurred. After adjusting for trend, seasonality and autocorrelation, a strong correlation was found between the two metrics of total antimicrobial use (r=0.68, p=0.001,Table1). The linear regression model for these two metrics use was: coefficient β=0.66 (0.45-0.88), p<0.001. Correlation between DDDs/100BD and DOTs/100BD varied among commonly used antimicrobial agents.

Table1: Correlation betweenDDDs/100BD and DOTs/100BD
Antimicrobial	Pearson's r	r 95% Cl	R squared
Clarithromycin	0.94	0.86-0.98	0.89
Ciprofloxacin	0.86	0.66-0.94	0.74
Amikacin	0.81	0.56-0.92	0.65
Clindamycin	0.78	0.51-0.91	0.61
Metronidazole	0.77	0.49-0.91	0.76
Meropenem	0.76	0.47-0.90	0.58
Teicoplanin	0.74	0.42-0.89	0.63
Piperacillin/ Tazobactam	0.62	0.23-0.84	0.71
Colistin	0.58	0.17-0.82	0.69
Vancomycin	0.54	0.12-0.80	0.67
Ceftriaxone	0.46	0.01-0.75	0.35
TOTAL ANTIMICROBIALS	0.68	0.33-0.87	0.72

Conclusions/Learning Points:

Use of a linear regression model to estimate DOTs/100BD from DDDs/100BD could be a useful approach for PICUs where access to patient medical records is not feasible, although further validation of the model is needed.

Hide