[Outcome measures in perinatal medicine--pO2 and SO2. With remarks on pulse oximetry]

Abstract

Background: Due to the outstanding pioneer work of Ronald E. Myers (Bethesda, Maryland) using term rhesus monkey fetuses we know for sure that hypoxia is the leading cause for brain damage and death when exposure occurs perinatally. He defined threshold values for oxygen content and time variables leading to death or cerebral injury. Years later pulsoximetry was developed for measuring fetal oxygen saturation (%) continuously. In this context the obstetrician wants to know: 1) what is the diagnostic potential of pO(2) (mmHg), SO(2) (%) and oxygen content (vol%) in umbilical blood? and 2) using these data could we ascertain fetal pulsoximetry which in addition uses the factor time.

Methods: In a sample of 7814 term fetuses, delivered in cephalic presentation by the vaginal route, actual blood gases and the variables of the fetal acid-base balance were determined in umbilical blood using equipments (BMS up to ABL 3) from RADIOMETER, Copenhagen. Measurements were done immediately post-partum by trained medical personal. Fetal oxygen saturation (%) for HbF was computed using the algorithm of Ruiz et al. Oxygen content (vol%) was determined according to Severinghaus using Hb values (g%) in each case.

Results: The median pO(2) in blood of the umbilical artery (UA) was 17.9 (mean: 18.8 +/- 8.3) mm Hg and in the umbilical vein (UV) 28.5 (mean: 29.3 +/- 9.2) mmHg, respectively. The median oxygen saturation (%) amounted to 24.8 (UA) and 60.9 (UV) using pO(2), pH and pCO(2) for computation in each case. The oxygen variable pO(2) showed no clinically important correlation neither with actual pH (r = 0.032, P = 0.005) nor with base excess (r = 0.047, P < 10 ( - 4)); the correlation with oxygen saturation (%) however was highly significant both for pH and base excess. Computational evaluation of the BOHR/HALDANE effect reveals that pH values of 6.9 are associated with 10 % oxygen saturation only in UA. P (50) values change dramatically with lowering pH values, i. e., acidosis. Oxygen content (vol %) offers no diagnostic advantages over oxygen saturation. Separation of the whole sample (N = 7814) according to the boundary of 30 % saturation in UA/UV blood leads necessarily to a big difference in saturation (DeltaSO (2)) but to an only very small difference in pH (DeltapH) and base excess (DeltaBE); thus the discriminatory power of SO(2) (%) is weak and clinically insignificant.

Conclusions: For fetal surveillance only oxygen saturation (%) -- not pO(2) and oxygen content -- is a useful variable. Due to the BOHR/HALDANE effect the fetus is able to rapidly mobilize his oxygen reserves and thus to compete with hypoxia; acidosis makes sense in this context. The three oxygen parameters analysed so far offer less diagnostic power than actual pH and base excess.