SAN DIEGO – Neonatal intensive care units are famously spendthrift when it comes to antibiotic utilization, but a well-executed antibiotic stewardship strategy can safely reduce unnecessary prescribing – as demonstrated at the 90-bed, level-IIIC neonatal intensive care unit (NICU) at Parkland Memorial Hospital, Dallas.
Results of the Surveillance and Correction of Unnecessary Antibiotic Therapy (SCOUT) study showed that total antibiotic days of therapy (DOT) in the Parkland NICU dropped by 27% following implementation of an antibiotic stewardship program featuring hard stops built into the electronic medical record, Dr. Joseph B. Cantey reported at the annual meeting of the Pediatric Academic Societies.
Dr. Joseph B. Cantey
This overall reduction in antibiotic utilization was accomplished through improvements in the three specific categories of NICU antibiotic use targeted for intervention in the SCOUT study: treatment courses of more than 48 hours for “rule-out sepsis” and treatment lasting longer than 5 days for pneumonia or “culture-negative sepsis.”
The proportion of antibiotic treatment courses for rule-out sepsis that were discontinued by 48 hours when cultures were sterile tripled from 32% to 95% between a 9-month baseline period and a second 9-month period after implementation of the antibiotic stewardship program. The proportion of courses of antibiotics for pneumonia that were limited to 5 days doubled from 36% to 72%. Similarly, there was a doubling in the proportion of treatment courses for “culture-negative sepsis” limited to 5 days, with the rate going from 31% to 62%, said Dr. Cantey, a pediatrics fellow in training at the University of Texas Southwestern Medical Center, Dallas.
During the 9-month baseline surveillance period, in which there were 1,607 patients in the NICU, the total antibiotic use was 343 DOT per 1,000 patient-days. During period two with the intervention in place, there were 895 NICU patients, and 251 DOT per 1,000 patient-days. The DOT, a commonly used measure in the field of infectious diseases, is determined by multiplying the number of doses by the dosing interval. DOTs in patients on multiple antibiotics are additive, he explained.
During the baseline observation period, 94% of all antibiotic use in the NICU was empiric therapy for suspected infection. More specifically, 63% of all antibiotic use was for rule-out sepsis. And while many of the courses of antibiotics given for this reason that exceeded the 48-hour limit during the baseline period did so by only one or two doses, those extra doses added up to 41 DOT per 1,000 patient-years, making this a worthy target for intervention. Together with treatment of pneumonia or “culture-negative sepsis” for longer than 5 days, these three high-yield targets accounted for 87% of all antibiotic use in the NICU during the baseline period, Dr. Cantey said.
The infants occupying the NICU during the two 9-month study periods were virtually identical in terms of their characteristics and reasons for admission.
Antibiotics are the most commonly prescribed medications in NICUs. Their use has been associated with adverse NICU outcomes, including increased risks of necrotizing enterocolitis, late-onset sepsis, and death in infants with birth weights below 1,500 g, as well as an increase in multidrug-resistant organisms, he noted.
In SCOUT, the composite outcome of necrotizing enterocolitis, late-onset sepsis, or death didn’t differ between the two study periods: 17.1% at baseline and 15.8% during the intervention period. Similarly, the incidence of colonization with multidrug-resistant organisms was 1.4% during the baseline period and not significantly different at 1% after implementation of the antibiotic stewardship program. Larger multicenter studies with pooled data will be required to determine whether antibiotic stewardship in NICUs affects neonatal outcomes, Dr. Cantey said.
He added that he and his coinvestigators are now trying to identify additional NICU scenarios in which antibiotics can safely be withheld. One likely candidate: asymptomatic preterm infants exposed to premature rupture of membranes. The investigators also hope to utilize the ongoing prospective surveillance element of Parkland’s NICU antibiotic stewardship program to identify the safe minimum treatment duration for common conditions such as urinary tract infections, sepsis, and necrotizing enterocolitis.
Audience members were effusive in their praise of the SCOUT study and the Parkland program. They wanted to hear more details about how the physician behavior change was accomplished. Dr. Cantey said that the three intervention targets were approved by the NICU medical director and the plan was disseminated to all the neonatologists, nurse practitioners, fellows, and residents. It was important to be able to assure everyone that outcomes would be prospectively monitored closely to ensure safety. The toughest task, he added, was to create the hard stops in the electronic medical record so that, for example, treatment for rule-out sepsis would automatically stop at 48 hours: skilled information technologists were essential.