Prospective Surveillance of Antibiotic Use in the Neonatal Intensive Care Unit

Results From the SCOUT Study

Joseph B. Cantey, MD; Phillip S. Wozniak; Pablo J. Sánchez, MD


Pediatr Infect Dis J. 2015;34(3):267-272. 

In This Article


During the 14-month study period, 1607 infants were admitted to the PMH NICU; 593 during the 5-month retrospective period and 1014 during the 9-month prospective period. Data were available on all infants, and there were no meaningful differences between the 2 periods (Table 1). The 1607 infants were administered 1420 antibiotic courses for a total of 9165 DOT [343.2 DOT/1000 patient-days (PD)] and a total LOT of 5215 days (195.3 LOT/1000 PD). All but 4 infants (99.8%) received at least 1 dose of an antibiotic. Seventy-two percent (n = 1157) of infants received 1 (n = 691, 43% of cohort) or more (n = 466, 29%) courses of antibiotics. The remaining 450 (28%) infants received intramuscular penicillin prophylaxis only. The 4 infants (0.2%) who did not receive antibiotic therapy had critical heart disease and were transferred immediately after birth to the cardiac intensive care unit at Children's Medical Center Dallas.

In all cases, the indication for antibiotic therapy was clear from review of the EMR. Before the beginning of the study period, the neonatology division had adopted problem-oriented clinical notes.[29] Every infant has a problem list maintained in the EMR, and the diagnosis made by the neonatologist as well as the antibiotic therapy prescribed is detailed in his or her daily progress note.[30] Of the 1420 antibiotic courses, 1116 (79%) were initiated within 72 hours of birth due to concern for early-onset sepsis. The remaining courses were initiated at ≥72 hours of age for suspected late-onset sepsis (21%) or for perioperative prophylaxis (0.2%). Gentamicin, ampicillin and oxacillin accounted for 93% of all antibiotic use (Fig. 1). Vancomycin and piperacillin–tazobactam accounted for 1% and 0.7%, respectively, of all antibiotic use. Meropenem (0.13%), cefotaxime (0.05%) and first-generation cephalosporin agents (0.04%) were the least-used antibiotics.

Figure 1.

Antibiotics used by DOT per 1000 PD.

Antibiotic use for specific indications is shown in Figure 2. Ninety-four percent of all antibiotic use (323 DOT/1000 PD, 154 LOT/1000 PD) was initiated as empiric therapy for suspected infection, 5% (16.6 DOT/1000 PD, 33.2 LOT/1000 PD) was penicillin prophylaxis for GBS and 1% (3.5 DOT/1000, 4.6 LOT/1000 PD) was perioperative prophylaxis for 49 surgeries on 41 infants. For empiric antibiotic therapy, 5% were for suspected infections that eventually were confirmed by positive cultures (Fig. 3; Gram-negative bacteria, 8.5 DOT/1000 PD; Gram-positive bacteria, 8.7 DOT/1000 PD). All infants treated for coagulase-negative Staphylococcus sepsis had ≥2 positive blood cultures.

Figure 2.

Flowchart of all antibiotic DOT (N = 9165) per 1000 PD administered to infants during the study period.

Figure 3.

DOT per 1000 PD used to treat pathogens causing culture-proven infection.

The remaining antibiotic use was in infants who had sterile cultures. "Ruled-out" sepsis—situations where cultures were sterile at 48 hours and antibiotics were discontinued—accounted for 63% of all antibiotic use (216.2 DOT/1000 PD, 94.1 LOT/1000 PD). Of these "ruled-out sepsis" courses, 32% were stopped within 48 hours, but 68% were extended beyond 48 hours despite physician documentation of intent to stop therapy. Prolonged (≥5 days) use despite sterile cultures accounted for 26% of all antibiotic use (89.4 DOT/1000, 46.8 LOT/1000 PD). The most frequent reasons for such prolonged therapy were pneumonia (16%) and "culture-negative" sepsis (8%) with length of treatment being ≥7 days in 64% and 69% of cases, respectively.