Doctors who use expensive, high-technology methods to save extremely premature newborns weighing less than two pounds must consider whether such treatment is justified given the suffering involved and the possibility of severe, permanent handicaps, two influential medical ethicists say.

In an article published recently in the American Journal of the Diseases of Children, Ernle W.D. Young, co-director of Stanford's Center for Biomedical Ethics, and David K. Stevenson, chief neonatologist at Stanford University Medical Center, recommend that such agonizing decisions be made on an individual basis by the parents and physicians and not by government fiat. The article is the latest round in a debate ignited in 1982 when the Reagan administration issued Baby Doe regulations, later overturned by the Supreme Court, requiring that hospitals treat handicapped infants over the objections of their parents.

William E. Benitz, a Stanford University neonatologist, said he agrees with Young and Stevenson. "We really do have an obligation to think carefully about whether these different interventions offer something to these infants, and if they don't, we shouldn't apply them simply because we have them," he said. "Because all of them are associated with some potential to do harm." Instead, he and the authors said, the medical profession should concentrate on reducing the number of premature and seriously ill newborns through improved prenatal care.

A study published in the July issue of Pediatrics, the journal of the American Academy of Pediatrics, compared two groups of very low birthweight babies -- less than 1 pound, 8 ounces. The first group was born between 1977 and 1980 and the more recent, between 1983 and 1985. It found that there were more survivors in the current group, but that the severity of their handicaps appeared to have increased. The study found that children delivered by Caesarean section did better as a group than those delivered vaginally and that the outlook for low-birthweight twins was extremely poor.

Whether the advances in sophisticated medicine that have enabled doctors to save progressively smaller newborns should be applied universally is, the Stanford specialists acknowledged, a subject of as much passionate debate among doctors as the more global question of health-care rationing is in society at large.

"You have these neonates born at very low birthweights being maintained on life support where they have, in essence, no chance to gain any quality of life," said Thomas Raffin, a specialist in respiratory medicine and co-director with Young of the Center for Biomedical Ethics. "Even if they survive -- an unlikely probability -- they will be so catastrophically impaired that their life will not be worth living."

One view, represented by some at Stanford, holds that it is more ethical to withdraw treatment from this group of newborns. The opposite opinion holds that because a newborn is incapable of conceptualizing "quality of life," any infant with any chance of survival should receive every available treatment.

In their article, Stevenson and Young suggested that some doctors, influenced by the legacy of Baby Doe, the fear of malpractice suits or both, are choosing heroic measures that may only buy a few years of painful, technologically assisted existence costing hundreds of thousands of dollars.

The Baby Doe legacy, Young and Stevenson wrote, consists of laws passed by some states to replace the Reagan administration's federal rules. Local statutes, many of which remain on the books, suggest that failure to treat could be interpreted as child neglect or abuse.

But as Young and Stevenson noted, the lives of many of these babies are short and dismal. A recent study conducted at Stanford showed that 35 percent of such infants survived; a 1986 study of severely premature babies conducted at six hospitals showed that one third of those who survived had significant "neurodevelopmental" handicaps. The cost of saving each of these infants, who were hospitalized an average of about five months after birth, was nearly $159,000.

"The extremely premature, very low-birthweight infant inevitably carries the care-giving team and the parents alike into a zone of ambiguity," they wrote. "Here, as is the case in any genuine moral quandary, the choices may not be between absolute 'rights' and {a human life} so much as between greater and lesser goods and harms."

"The effects of erring on the side of life are as follows," they continued. "We save some who would otherwise have died; we do immediate harm and inflict long-term suffering on many who survive, and we expend an enormous amount of money on neonatal intensive care."

Stanford's Benitz cited a French proverb for healers that he translates as "to comfort, always; to palliate, often; to cure, occasionally." Until the last century, he observed, most of what physicians did was "hand holding and comforting patients and their families. That's not a tradition we should lose. We tend to forget," he said, "now that we're curing so much more often than in the past, that the other things are also very important."

The atmosphere of an intensive-care nursery is such that heroic interventions, miraculous for some, may mean prolonged torture for others. "Lay people and even medical people unfamiliar with the environment may not understand that," Benitz said.

"We can't justify doing things that damage babies if those same things don't clearly offer the hope of helping," he said. "We have to focus on what is best for the specific baby." Extremely premature births, he notes, are to a large extent preventable.

Rae Grad, executive director of the National Commission to Prevent Infant Mortality, agreed. "We don't make a judgment about who should be saved and who shouldn't," said Grad. "What we say is that you wouldn't get yourself into this quagmire if you put the dollars into prevention. The data is so strong to show that if you did, you wouldn't have ethicists and neonatologists worrying about this."

A report issued recently by the commission entitled "Troubling Trends: The Health of America's Next Generation," called low birthweight the primary factor in infant mortality and subsequent disability. Preventable causes cited were poverty, lack of access to health care, smoking, alcohol and drug use during pregnancy, poor nutrition and teen pregnancies. The report noted that inadequate nutrition alone "may account for as much as 57 to 65 percent of babies born with low birthweight."

A report by the Office of Technology Assessment, the congressional research organization, found that "for every low-birthweight birth averted, we save between $14,000 and $30,000 of the associated health care costs."

"I wouldn't mind being put out of business if it meant that babies would be born healthy instead of sick," said Benitz. But, he added, "in our current political climate, I'm not worried about that."