Resident Johns Hopkins All Children's Hospital Tampa, Florida, United States
Background: Hyperoxia (SpO2 >95%) has shown to increase the rate of ROP and lead to adverse pulmonary effects. However, lower oxygen saturation goals (SpO2 of 85-89%) were also shown to increase the rate of mortality and necrotizing enterocolitis, as compared to an oxygen saturation (SpO2) of 90-95%. Despite the known importance of targeting a narrow SpO2 range, in the JHACH NICU very low birth weight (VLBW) infants are only within the recommended goal range of 90-95% 45 percent of the time. We hypothesize that use of histograms, accurate alarm parameters and integration of this data into the electronic medical record (EMR) can help increase time at the target SpO2 range. Achieving SpO2 in the target range more frequently will lead to safer and more effective care and may ultimately lead to decrease in the rate of ROP. Objective: To improve the duration of time spent in the target range for oxygen saturation (SpO2 90-95%) from 45% to 65% in VLBW infants in the NICU by March 2024. Design/Methods: The study was conducted at Johns Hopkins All Children’s (JHACH) hospital, a tertiary academic pediatric center with a 97-bed level IV NICU. The project was acknowledged by the JHACH institutional review board. A Key Driver Diagram was developed at the start of the project. Major drivers included histogram knowledge and awareness, utilization of histogram data, integration of histogram data into the EMR, and accurate alarm limits. Based on these drivers 4 PDSA cycles were designed for implementation between April 2023-March 2024. Inclusion criteria was birth weight < 1500g. Exclusion criteria included infants with PPHN and cyanotic heart disease. The primary outcome was average time spent at goal SpO2. SpO2 ranges were assessed by random auditing of the histogram on bedside cardiorespiratory monitors. Baseline data was obtained 3 months before the start of the project. As process measures, percent of histograms correctly documented in EMR per 24 hours and accurate alarm limits were tracked. For a balancing measure, time spent below target (SpO2 < 95%) was used.
