RESUMO
Antibiotic overuse in infants is associated with an increased risk of serious adverse events. Development of antibiotic stewardship programs aimed at reducing overall antibiotic consumption requires epidemiological surveillance. Retrospective surveillance and evaluation of all antibiotics provided to every infant admitted to maternal wards or neonatal intensive care units (NICUs) from 01 January 2014 to 31 December 2014 were performed in five medical centers of Saint Petersburg, Russia. Types of antibiotics and dates of administration were recorded. Antibiotic use was quantified by length of therapy (length of therapy, LOT, per 1000 patient-days, PD) and days of therapy (DOT/1000 PD). An additional parameter named "instant DOT/1000 PD" was introduced by authors for assessment of longitudinal patterns of administrations. Antibiotic load was 825.6 DOT/1000 PD in maternity wards and 1425.8 DOT/1000 PD in the NICUs. These levels are two to four times higher than DOTs reported in the USA for a level III NICU (348 DOT/1000PD). Antibiotic load was associated with the length of hospital stay (LOS) and birth weight. These associations were distorted when assessed using the conventional parameters, LOT and DOT, because they do not reflect the longitudinal component of treatment and underestimate antibiotic load when a patient stays in hospital without treatment. The proposed additional parameter successfully overcame these flaws and uncovered hidden associations. Severe overuse of antibiotics may be taking place in Russia and antibiotic stewardship development should be urged. Instant DOT/1000 PD is a more powerful tool in assessing treatment patterns than DOT/1000 PD.