The emergence of antibiotic resistance has contributed to the development of multidrug-resistant bacteria, which is a major concern.

The primary objective was to explore the possible association between antibiotic use and the emergence of resistance in a mother-child university hospital.

This retrospective study was conducted in a university hospital centre. Antibiotic-bacteria pairs were established, taking into account the number of isolates, actual antibiotic use, and clinical relevance. For each pair, a comparison of 2 variables (antibiotic utilization and rate of resistance) was quantified with the Pearson coefficient. Three analyses were conducted: no lag between utilization and resistance, 1-year lag, and 2-year lag.

Thirty antibiotic-bacteria pairs were selected from hematology-oncology and 18 from neonatology. In hematology-oncology, 6 pairs had a positive correlation (Pearson coefficient > 0.7): 2 pairs involving meropenem, 2 involving ceftazidime, and 2 involving piperacillin-tazobactam. In 3 of these cases, there was no lag between consumption of antibiotics and presence of resistance. In neonatology, 3 antibiotic-bacteria pairs had a positive correlation, 1 each involving vancomycin, cloxacillin, and meropenem.

It is possible to explore the potential association between consumption of antibiotics and emergence of resistance in a particular centre. Our exploratory approach was based on manual data processing. It would be interesting to consider a continuous systematic approach, allowing automatic generation of correlations.