Patient autonomy is considered central to modern medical ethics. In the absence of a competent patient, surrogates ought to make decisions regarding invasive medical procedures. For newborns, the natural surrogates are parents. We wondered to what extent parental exercise of surrogate autonomy actually impacted decisions to provide resuscitation in the delivery room for extremely premature newborns whose viability and quality of life were, at least statistically, uncertain.

Methods: We surveyed 550 neonatologists regarding their delivery-room resuscitation decisions in 4 potential scenarios: 1) birthweight (b.w.) < 500 gms/ gestational age (g.a.) < 23 wks;2) b.w 500 - 600 gms/ g.a. ≈24 wks; 3) b.w. 600 -750 gms/ g.a ≈25 wks; 4) b.w. > 750 gms/ g.a. 26+ wks). For each scenario, we asked whether the physician would choose a) full resuscitation; b) comfort care; c) defer to parents wishes. We also asked the physicians to rate which, if any of the following factors affected their decision: viability, futility, quality of life, resource allocation, fear of litigation, religious beliefs.

Results: We received 362 completed responses (66% response rate). For b.w. < 500 gms, 58% of respondents chose comfort care, 35% chose to defer to parents wishes, and 6% chose full resuscitation in the delivery room. In contrast, for infants 500 - 600 gms b.w. 65% chose full resuscitation, 34% chose to defer to parents wishes, and 1% chose comfort care. For infants 600-750 grams b.w., 92% chose full resuscitation, 8% chose to defer, and 0% chose comfort care. For infants > 750 grams b.w., 100% chose full resuscitation. In all birthweight groups and for all decisions, viability/futility/quality of life were deemed very important, while resource allocation/fear of litigation/physician's religious beliefs were not.

Conclusions: 1) When making decisions regarding delivery-room resuscitation of extremely premature infants, physicians placed great emphasis on patient-oriented outcome variables, while de-emphasing societal or personal concerns. 2) For no birthweight/gestational age category did a majority, or even a plurality, of neonatologists allow parental preference to determine delivery room decisions regarding resuscitation. 3) Parental exercise of surrogate autonomy, even for obviously burdensome care with low probability of a good outcome, remains a minor influence in decisions regarding delivery room resuscitation of extremely premature infants.