乌鸦传媒

A change in treatment guidance to reduce the use of vancomycin in the neonatal intensive care unit (NICU) for late-onset sepsis was safe, sustained, and resulted in similar mortality outcomes, a study of three Ohio NICUs found.

In the retrospective review of over 350 infants who received an empirical antibiotic course for possible late-onset sepsis, use of vancomycin fell from 84% in 2013-2014 to 25% in 2017-2019 following vancomycin reduction guidelines for those without a history of methicillin-resistant Staphylococcus aureus (MRSA).

And the proportion of infants receiving nafcillin jumped from 16% to 75% between the two study periods, reported Jacqueline Magers, PharmD, of Nationwide Children's Hospital in Columbus.

Despite the change, in-hospital mortality was not significantly different before or after the guideline change (9% vs 10%; OR 0.97, 95% CI 0.40-2.34), according to their findings in .

"Nafcillin can be a safe alternative to vancomycin for empirical therapy of LOS [late-onset sepsis] among NICU infants who do not have a history of [MRSA] infection or colonization," the study authors wrote.

Two infants in the study had antibiotic therapy restarted for recurrence of an infection 14 days after discontinuing initial therapy with vancomycin, the researchers said. No infant who received nafcillin empirically or for definitive treatment restarted antibiotics for a recurrence of the same infection.

In 2014, the Neonatal Antimicrobial Stewardship Program team at Nationwide Children's Hospital recommended that physicians use nafcillin instead of vancomycin for empirical therapy of possible late-onset sepsis (3 days after birth) in newborns with no history of MRSA. Those with known colonization with MRSA were treated with vancomycin.

As more preterm infants survive in the NICU, the rate of late-onset sepsis . The guideline change stemmed from concerns that widespread vancomycin use could lead to resistance in gram-positive bacteria, including coagulase-negative staphylococci (CoNS), which is .

"Most antibiotic usage in the NICU is for empirical therapy," the investigators noted. "Therefore, our strategy to reduce overall vancomycin usage targeted initiation and not just discontinuation of initial therapy when cultures did not yield a pathogen only susceptible to vancomycin."

The study was a retrospective review of all infants who received treatment for possible late-onset sepsis at three NICUs associated with Nationwide Children's, which included 516 evaluations for sepsis among 366 infants -- 113 during the pre-intervention period (2013-2014) and 253 during the post-intervention period (2017-2019). Of these, 78% discontinued antibiotic therapy within 48 hours and the remaining were definitive therapy.

The two time periods analyzed in the study allowed for enough time for the guideline changes to take effect, the researchers said.

Most infants were preterm, with a median gestational age of 28 weeks and median birth weight of about 1 kg. Overall, 57% of the infants were boys, and 10% died during NICU hospitalization. The number of infants who had a central venous catheter during sepsis evaluation was similar in the pre- and post-intervention periods.

Of note, during the post-intervention period there were higher rates of colonization with MRSA compared to the pre-intervention group among infants who underwent subsequent sepsis evaluations (18% vs 11%). In line with the guidelines, more of these infants received vancomycin.

Fifteen infants who received nafcillin were followed up with vancomycin: seven of those infants had proven infection with CoNS; five had no identified pathogen and received prolonged therapy; and two were showing lack of clinical improvement, resulting in a change in therapy. On the other hand, 13 infants who received vancomycin were changed to nafcillin treatment: five had methicillin-susceptible S. aureus bacteremia; two had methicillin-susceptible CoNS bacteremia; and six were without an identified pathogen.

In blood isolate analysis, 58% (23 of 40) of CoNS isolates were determined to have caused a proven infection, and 91% (30 of 33) were methicillin-resistant. In addition, of 16 infections due to S. aureus, 13 were susceptible to nafcillin.

The researchers noted a significant limitation to their study was that they could not account for other changes in infection prevention practices during the 7-year period, such as the institution of the Neonatal Antimicrobial Stewardship Program committee and the implementation of central line-associated bloodstream infection bundles.

Comment