REMOTE UMBILICAL ARTERIAL BLOOD-GAS ANALYSIS - EXPERIENCE AT 2 COMMUNITY HOSPITALS

Objective: Regression equations permit determination of umbilical arte rial acid-base analysis hours after delivery, and this is most benefic ial at community hospitals. The purpose of this investigation is to as sess the relative accuracy of remote umbilical arterial blood gas anal ysis at two community hospitals.Study design: For a month two health c are providers obtained two umbilical arterial samples after each deliv ery. While one sample was analyzed within GO mill of delivery, the oth er syringe was refrigerated in ice, and the acid-base measurements wer e made at variable time intervals. Student's t test, Chi square, or Fi sher exact lest were used where appropriate, and P < 0.05 was consider ed significant. Results: Both the incidence of neonatal acidemia (13/4 3 versus 3/35, P = 0.04) and the time interval from delivery to remote analysis (46.2 +/- 12.8 versus 31.9 +/- 14.4 h, P < 0.0001) were sign ificantly different at the two hospitals. Yet, at the two hospitals, t he calculated acid-base analysis had a similar ability to accurately i dentify whether a newborn was acidotic at birth (33/43 versus 26/35, P = 1.00). Moreover, at the two locations, use of mathematical equation s and raw data from the remote analyses predicted similar incidences o f pH greater than or equal to 7.20 and respiratory and metabolic acide mia as did the original umbilical arterial blood analyses (P > 0.05 fo r all comparisons). At the two centers, for the calculated pH < 7.20 t o identify a newborn with acidemia, sensitivity, and specificity, posi tive and negative predictive values were 92-100, 70-72, 25-58, and 95- 100%, respectively. Conclusion: It is feasible, with reasonable accura cy, to utilize remote umbilical arterial blood gas analysis at communi ty hospitals.