DILEMMA: You're an experienced orthopedic surgeon who has determined that your patient needs an extensive operation on his hip. You're aware that surgery, even under the best conditions, sometimes results in minor cuts and abrasions to your fingers. But you suspect that your patient is a homosexual (or intravenous drug user), so you request that he undergo blood testing for the AIDS virus. He refuses. You discuss the problem with a pathologist colleague, who suggests: "Draw a blood sample on him and tell him it's a 'routine test' and send it to me under a fictitious name. I'll call you with the result."

Question: What do you do?

Hint: For the moment ignore the very real possibility that your patient, if he learns of the deception, is likely to sue you.

As the number of AIDS cases increases dramatically, physicians are increasingly likely to come into contact with patients who are HIV positive. Surgeons are particularly at risk, since the virus can be transmitted via blood-to-blood contact. But what to do when the patient refuses testing?

The British Medical Association already has answered. At a recent meeting, by a 183-140 vote, members decided that "testing for HIV antibody should be at the discretion of the patient's doctor and should not necessarily require the consent of the patient." As one London anaesthetist put it, "I, for one, feel my life and those of my colleagues are more important than the future employment and insurance prospects of infected individuals."

On pragmatic grounds alone it's hard to fault that reasoning. Granted, it's not particularly magnanimous. But nobody wants to contract a deadly incurable disease, indeed, would do anything possible to avoid doing so. Doctors are no different from anybody else here. The question is: Should they be. (On Dec. 1, surgeons from around the world will address that issue. They will meet in London for a conference titled "AIDS and Surgery." Participants will discuss the implications of AIDS for surgical and associated medical procedures, and explore policies for the future.)

Over the centuries physicians have generally accepted the risks involved in caring for patients with contagious diseases. Cholera, yellow fever, tuberculosis, influenza -- each has claimed the lives of some of those who ministered to its victims. But the risk of death, generally a small one, has always been considered part of the responsibility a physician freely undertakes when he enters the profession.

However, most of the victims of AIDS are perceived as having become ill as a result of their own behavior. This is far different from the familiar paradigm, in which a person contracted an infectious disease through "no fault of his own," except to be in the wrong place at the wrong time. Even plagues and pestilences are associated in the minds of most physicians with sick people who, except for their illness, are no different from themselves. But AIDS isn't like that.

In a study published in the July issue of the American Journal of Public Health, researchers from the University of Mississippi reported that physicians tend to stigmatize AIDS patients to a "disquieting" degree. The researchers sent four variations of a fictional case history to 500 randomly selected doctors in Memphis, Phoenix and Columbus, Ohio. In every case, the history described a young man employed in the computer field who was suffering from fatigue, repetitive infections and general decline. In some versions, however, the patient was described as homosexual; in others, heterosexual. In some he was diagnosed as having AIDS, in others leukemia.

The result: When the patient was described as having AIDS, the doctors were more likely to say he was deserving of his fate, more dangerous to others and less deserving of sympathy and understanding. "It is naive to assume that all physicians will have equally positive attitudes towards all patients," the researchers wrote. "At the same time these data indicate that it is equally naive to assume that AIDS patients are viewed nonjudgmentally by physicians."

Although there are many reasons for this negative bias -- which no doubt extends beyond the cities studied, I believe it breaks down into two main categories.

First, the behaviors in question -- drug use and homosexuality -- are not "medical conditions" in the strict sense. No one claims that drug addiction can be cured or even held in check solely by medication or other primarily medical approaches. Homosexuality has been stricken from the list of mental illnesses and according to generally accepted psychiatric opinion is simply an alternative lifestyle.

Second, drugs and homosexuality conjure up, to varying degrees in different people, an emotional mixture of phobia, value judgment and avoidance. Add to this the fact that doctors on the whole tend to be a fairly conservative group.

Consequently, many physicians are put in the unique position of coming into daily contact with and caring for two population groups with which, on the whole, they would prefer not to associate.

To make matters worse, there is a risk of contracting the deadly disease from patients. Within the past year, several health-care workers, outside of the known risk groups, have become HIV (human immunodeficiency virus) positive -- presumably on the basis of inadvertent exposure while carrying out their duties. At this point it seems that the risk is small. But it is not negligible. Put differently, even if the risk of contracting the disease from an AIDS patient is one in a million, the risk of contracting it from caring for a patient who is HIV negative is zero in a million.

In the current climate of concern over privacy, as more legal strictures are put into place to guarantee "confidentiality" and protect patients' "civil rights," the more likely it is, I believe, that physicians, for their own protection, are going on resort to indirect, surreptitious and -- let's call it what it is -- illegal ways of getting the information they need to protect themselves. Furthermore, technology will soon be available to make these under-the-table determinations exceedingly easy to carry out.

In early 1988 Dupont plans to introduce into Great Britain the world's first instant AIDS test. It will require no more than a drop of diluted blood which can be placed within the center of a small packet about the size of a Band-Aid. The color in the central well of the packet changes according to whether antibodies to HIV are present or not in the patient's blood.

Equipped with such a device, a doctor, after withdrawing a "routine blood sample" could, if he so desired, place that final drop of blood -- the one that's customarily blotted off the skin with a cotton swab -- into the central well of the "test kit" concealed in the palm of the hand. There are no samples to be sent to laboratories, no codes to be entered, no forms to be filled out, no questions to be answered, no "red tape" and, most importantly, no practical way of being detected. Although it's true -- as DuPont freely admits -- that the test yields a small number of false positives, as a screening procedure it could hardly be better.

If all of this sounds fantastic or exceedingly improbable, remember that at least some doctors consider it a life-and-death issue. To most people outside high-risk groups, the chance of getting AIDS seems fairly remote; but they may seem far less remote to a surgeon or other physician who routinely deals with blood and other potentially contaminated fluids.

"When I chose medicine as a career I never signed on for anything like this. I am looking out for No. 1 here!" Is there any physician in this country to whom this thought hasn't occurred? But if surreptitious AIDS testing becomes a reality -- as I believe it will -- additional dilemmas are likely to arise.

Suppose, in the example cited above, the HIV test comes back positive? What then? When the surgeon tells his patient that he's not going to perform the operation, does he also tell him that his results came out positive in a test for which no permission was ever obtained? Not likely.

Does he destroy the test results? To do that is to engage in an act even more egregiously irresponsible than he believes his patient to be in placing his own "civil rights" before the surgeon's safety. The patient didn't actually know that his condition when he refused to test; he simply didn't want to find out. The surgeon, however, now knows that the patient is positive and must decide what steps he can and should take to protect others.

Into this cauldron of suspicion, anxiety and, on occasion, outright paranoia, drop in a pinch of litigiousness. For reasons that will no doubt interest future historians, a goodly percentage of the population is still more concerned about "civil rights" than it is about people dying. So the doctor caught secretly carrying out an AIDS test is likely over the next several years to spend more time in depositions than in operating theaters.

What would you do if faced with such a situation?

"I'd take extra precautions with such a patient who was unwilling to undergo HIV testing. But I wouldn't refuse to perform the operation," responds a surgeon whose courage and compassion I greatly admire.

But a more probable answer these days is: "No test, no operation. While it's true that you don't have to take the test, it's also true that I don't have to perform the operation. Here are the names of some other orthopedic surgeons in the area. Try your luck with them."

Does this strike you as harsh and unfeeling? Do you think such a doctor should have his license lifted? According to one theory about doctors and their obligations to the public, you would be considered correct.

In the face of what's yet to come with the AIDS epidemic, legislators and judges may soon be forced to decide whether doctors are free, as they have been traditionally, to treat patients according to personal choice, or whether a physician's unique position is no more than a privilege bestowed by society at large in the form of a license.

According to the latter theory, what society has granted, society can take away when physicians' practices fail to conform to society's expectations. From that perspective, doctors' fears of contracting AIDS from a patient might be judged simply irrelevant: "If a doctor is that worried about getting sick, maybe he should have chosen investment banking as a career."

On the other side, physicians, particularly surgeons, are asking, "Why should we be the only group in society that can be forced into engaging in high-risk behavior?" And they are understandably concerned that, for the most part, those who are most loudly defending civil rights are not the same people who are in the operating rooms taking the risks.

Looking back now, 20 years after graduation from medical school, I can empathize with the "I-never-signed-on-for-anything-like-this" orientation. At the same time I recognize my duty and privilege to administer to sick people who, at the very least, deserve to be spared my own personal views about the quality of their lives. But I also believe that doctors shouldn't be forced to resolve on their own issues that the majority of people in our society haven't decided about as yet.

Should HIV testing be made mandatory for members of high-risk groups who require medical attention? Should all patients admitted to hospitals be tested, thus circumventing prejudice or discrimination? At the moment, a no national policy exists. Until such a policy is formulated and a consensus is reached regarding how much time, effort and money will be expended towards the control and cure of AIDS (and no one, whatever his orientation, believes that we have done enough), the best that can reasonably be expected from physicians is the "no test, no operation" option.

It protects everybody's interest: The patient retains his privacy; the doctor doesn't have to pull a fast one that may result in his breaking the law; the public doesn't have to fear that their doctors are testing for anything other than what is printed on the laboratory slip. And finally, physicians will be free to choose -- for themselves and for their own reasons, logical or illogical -- whether or not they wish to place their lives in peril.

Richard Restak is a Washington neurologist and the author of "The Brain."