| Abbas Tavakoli, Isabelle Bennett, Robin Dail |
Monitoring of physiological vital signs from body temperature and heart rate is crucial to guide treatment for preterm
infants to optimize health outcomes. Monitoring continuous measurement of both central (abdominal) and peripheral
(foot) skin temperatures will give a thermal gradient, which may predict the onset of infection. Studies have shown that
heart rate characteristic (HRC) scores can be used to predict infection; however, a special HeRO monitor must be used in
addition to standard care. Our team compared thermal gradients to HRCs to predict infection in early preterm infants
(born < 32 weeks gestation age). 364 preterm infants were enrolled in the study after approval of IRB and parental
consent. Each infant had a skin temperature probe (thermistor) attached to their abdomen and another to the sole of one
foot. Temperature and HRC data were downloaded from a research datalogger, and HeRO monitor as text files, then
merged into Excel format, with each infant's data containing approximately 41,002 rows. Data were collected from five
different sites: Duke (n=25), ECU (n=137), MUSC (n=31), Prisma (n=149), and UNC (n=11). SAS datasets were created
for each infant, and a macro was used to combine the data from all infants at each site. Demographic variables were then
merged with the combined site data, and finally, the datasets from all sites were merged. Several macro statements were
written to facilitate the combination and merging of datasets using SAS, version 9.4. |