An exceptionally small virus that causes a common contagious childhood illness many people have never heard of is turning out to account for a variety of other health problems as well. Among them are some cases of arthritis, certain complications of genetic blood disorders like sickle cell disease, and injury to fetuses.

Erythema infectiosum, sometimes called "fifth disease," typically gathers steam in occasional school epidemics. Although the disease was identified almost a century ago, not until 1983 was it learned that an agent discovered only eight years earlier, called parvovirus B-19, is to blame. Knowledge of other mischief that B-19 is capable of is more recent still and by no means complete even now.

According to estimates from the Centers for Disease Control, there are 2 million to 3 million cases of fifth disease in the United States alone each year, mostly in children ages 5 to 14. The virus is occasionally spread by blood transfusions or by injection of therapeutic products derived from blood. But how one person usually gives it to another is unknown, though it is suspected that the infectious material travels in airborne secretions from the respiratory tract.

Fifth disease got its name around 1900, when medical textbooks assigned numbers to the contagious childhood illnesses accompanied by skin eruptions. Scarlet fever, for example, was "first disease," and measles was "second disease." Though modern texts have dropped this practice, the term fifth disease for erythema infectiosum has stuck.

Fifth disease is ordinarily mild, and some children get it without showing any symptoms. It most often announces its arrival by a fiery red facial rash that makes the cheeks look as if they had been slapped.

After that, the rash typically spreads to the trunk and limbs, where it gradually fades, assuming a lace-like pattern, which then comes and goes for one to three or more weeks. It is a temperature-sensitive rash that becomes more obvious on exposure to heat and may come back temporarily, once the child has recovered, after a hot bath or exposure to sun.

Most children with fifth disease hardly feel ill, although 10 to 30 percent have some symptoms -- such as slight fever, itching, sore throat and abdominal or joint pain -- as well as the telltale rash. Says Dr. Georges Peter of Brown University, head of the American Academy of Pediatrics' committee on infectious diseases: "Fifth disease is of no lasting consequence to the child."

It has thus come as somewhat of a shock to scientists that parvovirus B-19 infections are not always benign.

When, for example, there were simultaneous outbreaks in Britain of fifth disease in normal schoolchildren and severe compli- cations in patients with sickle cell anemia, researchers there were able to put two and two together and trace both epidemics to the same cause: B-19. This virus, it turns out, attacks the bone marrow, where red blood cells are made.

The bone marrow of healthy children is very little affected. But in children with hereditary anemias -- such as sickle cell -- the situation can be more serious. They may need heroic supportive measures to tide them over the consequent crisis.

People with weakened immune systems also may be vulnerable. The case, recounted in a recent issue of the New England Journal of Medicine, of a California boy who is now almost 3 is illustrative.

Born with an immune deficiency disorder called Nezelof's syndrome, the child suddenly became severely anemic and was eventually taken to the National Institutes of Health. There, Drs. Gary Kurtzman and Neal Young of the National Heart, Lung and Blood Institute found that his bone marrow was chronically infected with B-19, which explained his low red blood cell count.

Despite a variety of treatments, the child's bone marrow continues to be infected and he to be infectious. Kurtzman, Young and their colleagues suspect the boy may not be unique.

Accordingly, they have begun to look for the B-19 virus in the blood and bone marrow of others with damaged immune systems, such as AIDS patients, cancer patients on chemotherapy and patients who have had organ transplants and so must take immuno- suppressive drugs to prevent rejection of their grafts. Said Kurtzman in an interview, "I'd be surprised if we didn't find it {B-19} in some of them."

Meanwhile, he reports that he has himself had fifth disease and the arthritis that often goes with it in adults, which he describes as "extremely painful."

Also a concern is that although healthy children who get fifth disease generally take it in stride, healthy people who get it as adults are not necessarily as lucky. Two British reports, both published in the scientific journal Lancet in 1985, found that newly infected adults -- most but not all of them women and many without the characteristic facial rash -- very frequently developed severe arthritis lasting from several days to several months. The pain kept many of these patients home from work.

The link between B-19 and acute arthritis in adults has also been found in Japan and the United States. In the most recent U.S. study, Dr. Paul R. Joseph, a pediatrician on the attending staff at North Shore University Hospital in Manhasset, N.Y., followed the parents, teachers and other adult close-relative contacts of 13 children in eight families who came down with fifth disease.

Six of these adults exposed to infected youngsters in four of the families, Joseph reported in the New York State Journal of Medicine last November, developed not only a rash -- although not the "slapped cheek" rash seen in children -- but also either severe joint pain alone or joint pain with swelling and inflammation.

Joseph said in an interview that fifth disease has not been shown to cause chronic arthritis but added: "Further research is needed before this possibility can be ruled out." Meanwhile, the questions about fifth disease now most on the minds of physicians stem from the 1985 discovery in Britain that, like the rubella (German measles) virus, parvovirus B-19 can travel from the bloodstream of an infected pregnant woman to that of her unborn child.

Unlike the rubella virus, the erythema infectiosum virus has not been shown to cause birth defects. However, Pediatric Alert, a newsletter for doctors, recently reported a European case in which a woman who got fifth disease in early pregnancy elected to have an abortion and the fetus was found to have eye abnormalities like those caused by rubella, as well as inflammation of the heart and other muscles. Whether B-19 actually caused these abnormalities or whether they were a coincidence is unknown.

"All I think we can say at this point," said Dr. Henry H. Balfour Jr., professor of laboratory medicine, pathology and pediatrics at the University of Minnesota in a telephone interview, "is that if B-19 does cause fetal malformations, it's probably quite rare; almost certainly far rarer than was the case with rubella before there was a vaccine for that disease."

Instead, the evidence to date is that maternal infection with B-19 either kills the fetus or has no discernible effect on it at all. Some babies of infected mothers die before birth because the virus causes hydrops, a condition that prevents their bone marrow from producing red blood cells and makes their bodies fatally swell. (Hydrops can be detected prenatally by ultrasound examination.) The rest are normal at birth and have no traces of the virus in their tissues.

What is the likelihood that a woman infected with B-19 during pregnancy will miscarry or have a stillbirth? The short answer is that no one really knows because the connection between this virus and damage to the fetus was made so recently. Still, studies done in Scotland and published in several medical journals, including Lancet and the New England Journal of Medicine, suggest that the risk, although not zero, is small.

Dr. Larry Anderson, who directs similar studies being done in this country by the CDC, said the risk is "probably going to be less than one in 10, maybe one in 20. And that's for women infected with the virus in early pregnancy. For those infected later, it may be even less. About 95 percent of the time, the fetus is probably not affected."

A major reason for Anderson's optimism is that blood tests on women known to have been exposed to the virus during pregnancy -- they may or may not have any symptoms of fifth disease -- suggest that about half of them are already immune to it before they conceive. "By age 20 or 30, about half the population seems to be immune," Anderson explained. "If you are immune when you are exposed, you have nothing to worry about."

However, the blood tests Anderson relies on are strictly research tools. They are not available to practicing physicians whose adult patients want to know whether they are at risk.

That being so, why not safeguard women of reproductive age by quarantining kids with fifth disease symptoms? The answer is that this illness is thought to be contagious only in the early stages -- before the slapped-cheek rash appears 12 to 14 days after exposure to the virus. In other words, it is assumed that to isolate a child after symptoms appear would be futile. Still, several experts interviewed for this article believe, this view may change as there has as yet been no definitive study of how long B-19 can be infectious.

Thus, if further research finds that the virus damages fetuses more often than is now suspected, the solution to the problem may be the same one that has made birth defects from maternal rubella largely a thing of the past: a vaccine given to children but really aimed at protecting women of childbearing age.

Judith Randal is a medical writer in Washington.