Clinical

How to reduce NICU transfers for asymptomatic hypoglycemia

 

Key clinical point: Feed newborns at risk for hypoglycemia before checking their blood glucose levels.

Major finding: After making that and other changes, the transfer rate for at-risk infants at a major academic center fell from 17% to 3%, without an increase in rates of symptomatic hypoglycemia and adverse events.

Data source: Quality improvement project at the University of North Carolina at Chapel Hill.

Disclosures: The investigators had no financial disclosures. The work was funded by the National Institutes of Health.


 

AT PHM 2017

– At the University of North Carolina at Chapel Hill, many infants who would previously have been transferred to the NICU for asymptomatic hypoglycemia now are staying with their moms, thanks to a new protocol that holds off on blood glucose testing until infants are fed for the first time and glucose homeostasis can begin.

 

Not too long ago, the university realized it had a problem that’s probably familiar to other institutions: Its system to monitor newborns at risk for hypoglycemia – those born to diabetic mothers, or who are small or large for gestational age – put too many infants with asymptomatic hypoglycemia into the NICU when they didn’t really need to be there.

Nurse practitioners “were tired of transferring babies they felt were responsive to feeding and did not actually require NICU care,” and “a growing number of families were unhappy with being separated from infants that were well-appearing and feeding well at a time when moms were trying to establish breast feeding and bonding. There was frustration with our protocol,” which “seemed rigid and outdated,” said Ashley Sutton, MD, a pediatric hospitalist at the university.

Dr. Ashley Sutton, a pediatric hospitalist at the University of North Carolina at Chapel Hill.
Dr. Ashley Sutton
Under the old system, blood glucose was checked within an hour of birth whether the infant had fed or not, and infants were sent to the NICU if glucose levels were below 25 mg/dL; the protocol didn’t take into account the normal physiologic glucose nadir after birth, or allow enough time for the initiation of glucose homeostasis. While nursery staff waited for NICU personnel to arrive, “[We’d do] nothing, when moms were there with milk,” Dr. Sutton said at the Pediatric Hospital Medicine annual meeting.

To fix the problem, Dr. Sutton and others on a multidisciplinary team implemented the American Academy of Pediatrics’ 2011 guidelines for monitoring glucose homeostasis in late-preterm and term newborns at-risk for hypoglycemia, with an additional mandate to initiate immediate, continual skin-to-skin contact at delivery (Pediatrics. 2011 Mar;127[3]:575-9).

Under the new system, children are fed with either their mom’s or a donor’s breast milk within an hour of birth, and the initial glucose check comes at 90 minutes; infants are transferred if blood glucose remains below 25 mg/dL after the second feeding. After 4 hours of life, glucose levels below 35 mg/dL trigger an evaluation for symptoms, not necessarily an automatic NICU transfer.

Labor and delivery nurses also are empowered “to immediately feed the baby no matter what number [they are] seeing,” Dr. Sutton said at the meeting, sponsored by the Society of Hospital Medicine, the American Academy of Pediatrics, and the Academic Pediatric Association.

The efforts have made a difference. The transfer rate for at-risk infants has fallen from 17% to 3%, and skin-to-skin contact is initiated within the first hour of life in 64%, up from 45%. Feeding of at-risk infants within the first hour has increased from 43% to 61%, and the first glucose check comes at an average of 97 minutes. The number of unnecessary NICU transfers of at-risk infants has fallen sharply.

Meanwhile, there’s been no increase in sepsis evaluations, adverse events, readmissions, and the rates of symptomatic hypoglycemia.

Dr. Sutton and her colleagues had no industry disclosures. The work was funded by the National Institutes of Health.
   Comments ()

Next Article: