Developments in obstetric medicine during the past ten to twenty years have transformed the clinical status of the fetus.  Traditionally physicians have been trained to assess fetal condition by indirect methods: palpating the fetus through the maternal abdominal wall and uterus, measuring hormonal milieu through maternal urine and serum, estimating statistical risks from parental medical histories. While the skillful use of these methods could produce highly reliable clues to fetal health and development, the fetus itself eluded direct examination. Throughout pregnancy the fetus could not be known, but only approached inferentially and probabilistically. Until recently suspected fetal anomalies have been treated indirectly too, by therapeutically managing the maternal environment. Unable to interact with the fetus in clear distinction for its host, physicians conceptualized the maternal-fetal dyad as one complex patient, the gravid female, of which the fetus was an intergral part.
High-resolution ultrasonography and techniques for sampling fetal blood, urine, and other tissue have changed this conceptual scheme. These diagnostic tools penetrate the opaque environment and reveal the fetus to clinical observation in all its anatomical, physiological, and biochemical particularity. When anomalies are detected, in utero medical and surgical procedures are already beginning to offer alternatives to therapeutic delivery and neonatal treatment. The biological maternal-fetal relationship has not changed, of course, but the medical model of that relationship has shifted emphasis from unity to duality. Clinicians no longer look to the maternal host for diagnostic data and a therapeutic medium; they look through her to the fetal organism and regard it as a distinct patient in its own right.
What ethical implications flow from the fetus's transformation from inferred to observed entity? Unfortunately, legal developments have tended to preempt ethical exploration of the new two-patient obstetric model. Some physicians, assuming enhanced rights on the part of the fetal patient, have sought and obtained court orders to perform fetal therapies (notably cearean deliveries) without maternal consent.  Although few in number, these cases raise the possibility of a new standard of clinical practice with far-reaching implications for civil and criminal liabilities to physicians and pregnant women. With legal stakes so high, it is not surprising that ethical inquiry has been displaced. Yet in the absence of independent and thorough ethical analysis one cannot judge whether these developments are compatible with fundamental values of medicine and medical care, and so one cannot know whether physicians have responsibilities, individually or collectively, to promote or resist them.
Well-grounded in law and ethics or not, cases of court-ordered fetal therapy have set the agenda for debate, focusing attention on the question, What should the physician do when a pregnant woman refuses medical or surgical treatment recommended for the well-being of the newly individuated fetal patient? The two-patient problem for the physician is seen to begin at the point of maternal refusal and is framed as a conflict between values of fetal benefit and maternal autonomy. The medical recommendation precipitating refusal is, presumably, unproblematic. But that presumption requires examination. Inherent in any conceptual shift is the potential for equivocation between the old paradigm and the new. If the physician's recommendation of fetal therapy incorporates one-patient thinking about the maternal-fetal relationship, questions about maternal refusal of that recommendation may be spurious, resting on a logically illicit hybrid of one-patient and two-patient conceptual schemes.
We need, I think, to gain a fresh perspective on this issue by stepping back from the legal debate and considering in a systematic way how ethical guidelines for prenatal medical care are altered by transition to the two-patient obstetric model. …