Developing an objective method for analyzing vital signs changes in neonates during general anesthesia

Commonly used general anesthetics are considered to be neurotoxic to the developing rodent brain, leading to poor long-term outcome. However, it is unclear whether these rodent studies can be extrapolated to the human neonate. Given that anesthesia for urgent neonatal surgery cannot be avoided, it is vitally important to assess other factors that may impact neurological outcome following anesthesia and surgery.

The purpose of this study is to identify thresholds for detecting vital sign deviations, which may have the potential for affecting neurological outcome following anesthesia and surgery in neonates. These data may be suitable to identify targets for prospective quality improvement projects and guide future research for strategies to reduce detrimental neurocognitive outcomes.

A retrospective analysis of vital sign data was performed for neonates (age ≤28 days), undergoing noncardiac surgery over a 4-year period (2010-2013). Thresholds for detecting bradycardia, tachycardia, hypothermia, hyperthermia, hypertension, hypotension, hypocarbia, hypoxemia, significant changes in mean arterial blood pressure, and periods of high inspired oxygen concentration, were proposed. Selected chart review, to identify additional risk factors, and identify sources of data artifact, was performed for 224 cases.

Data from 435 procedures in neonates, with median (IQR [range]) ages of 6 (2-16 [0-28]) days were available for analysis. Five (3-6 [0-12]) rule deviations per case were observed; only 11 cases had no rule deviations. Hypothermia was observed in 285/435 (70%), moderate hypocapnia in 298/430 (69%), and severe hypotension in 270/435 (62%) cases.

An objective method of comparing cases has been created with a method to automatically identify neonatal vital sign deviations. With further validation the method has the potential to be a powerful tool to drive future quality improvement projects in neonatal anesthesia.