FLORENCE, Italy -- It was one of the most dumbfounding mysteries of the AIDS epidemic -- the case of the dentist who infected his patients. And a major study was launched to solve it.
This week, the disease detective who heads up that investigation delivered his findings.
Standing before an audience of more than 500 people at the Seventh International Conference on AIDS, Harold Jaffe, one of the most respected epidemiologists at the U.S. Centers for Disease Control, had to admit defeat. He simply could not explain how the dentist had infected five patients with acquired immune-deficiency syndrome.
The dentist, David Acer of Florida, would not cooperate with the investigation, Mr. Jaffe said. Some key documents unaccountably disappeared. People who might have the most vital information could not be found.
And when Dr. Acer died in September, he took a lot of secrets to his grave.
The case still is the only known one of a health-care worker transmitting AIDS to a patient. It raises urgent questions about the possibility that dentists, surgeons and other health-care workers with AIDS might be infecting their patients.
Mr. Jaffe, a quiet-spoken, unassuming man with little round eyeglasses and a concerned expression, delivered his conclusions Thursday without emotion, but with impressive authority -- despite the frustrating holes in the story he had to tell.
It all started 14 months ago, Mr. Jaffe said, when Kimberly Bergalis of Fort Pierce, Fla., tested positive for the virus that causes AIDS, known as HIV (human immunodeficiency virus).
Subsequent investigation linked her infection to Dr. Acer. In genetic tests, virus samples from the 22-year-old woman matched almost perfectly those from her dentist. Because no two AIDS viruses are identical, this strongly suggested that one person had infected the other, and because Dr. Acer's illness was far more advanced than Ms. Bergalis', presumably it was he who infected her.
At the time, it didn't look like much of a mystery.
HIV is nowhere near as easily transmitted as airborne microbes like the flu virus, which can be passed on with a sneeze, a kiss or the sharing of eating utensils. In the absence of sex or the sharing of needles, blood-to-blood contact would have been necessary for infection.
The most likely explanation, then, was that the dentist had cut himself while performing a tooth extraction, and some of his contaminated blood had entered the extraction wound.
This theory was seriously undermined a few months later, though, when Barbara Webb, a 65-year-old retired English teacher, contacted health officials. After reading about Ms. Bergalis, Ms. Webb had herself tested and found that she, too, had HIV infection.
Her virus was virtually identical to the one that infected Dr. Acer and Ms. Bergalis.
Could Dr. Acer have cut himself twice, and both times dripped contaminated blood into the dental wounds of his patients? Possible, but unlikely.
Then, health officials heard from another person, a young man this time, whom Mr. Jaffe referred to as Patient C.
Patient C was also HIV-positive, with no known risk factors other than having Dr. Acer as his dentist. Investigators tested the virus in his body. It matched the others.
Concerned that this might be the start of a one-man mini-epidemic, state health officials began sending letters to Dr. Acer's patients -- he had more than 2,000 in his general dental practice. Meanwhile, the CDC matched the list of patients with a list of known AIDS patients in the state.
One name was found on both lists. Patient D.
But that man's virus didn't match the rest, and he had other risk factors, making it unlikely that his infection was related to those of the first three patients. Citing privacy concerns, Mr. Jaffe refused to specify what the man's risk factors were.
Patient E, a woman, was identified in December. She, too, was a patient of Dr. Acer, and her virus matched his. Her lover also was one of Dr. Acer's patients, and tests showed that he too was infected with HIV, but his virus didn't match the others genetically. This suggested that his infection was in no way related to those of either his dentist or his lover.
And finally there was Patient G, a blood-plasma donor identified during routine screening. His dentist was Dr. Acer, and his virus matched the others.
Dr. Acer, then, had seven HIV-infected patients. And five had a virus that was almost identical to the one that was killing him.
All the patients had undergone tooth extractions or root canal work -- procedures involving sharp instruments that could cut tissue in the mouth, leaving it open to contamination with infected blood, Mr. Jaffe said.
Dr. Acer did not start using gloves until late 1987, when he was diagnosed with AIDS. But it was awfully hard to believe that -- five times -- Dr. Acer had injured himself so badly that he drew blood and that -- five times -- the blood had seeped into the wounds of his patients.
So if Dr. Acer did not bleed on his patients, what did happen?
One possibility -- though Mr. Jaffe admits that this, too, is implausible -- is that Dr. Acer's instruments were contaminated and spread the virus.
Good dental practice calls for changing instruments between patients, Mr. Jaffe said, so that anything coming in contact with the mouth is sterile. But some dentists become lax, perhaps failing to change the tip on a sucking device, or using the same probe on two patients without sterilizing it.
Still, if the virus were transmitted in this way, it means that each of the five patients had to have been directly preceded by patients infected with HIV. Deadly as it is, the virus isn't hardy enough to survive for many minutes in the fresh air on the tip of
an instrument. Even in fluids, Mr. Jaffe said, its survival time is just a few hours.
Interviews revealed that Dr. Acer was bisexual and that some of his lovers were also his patients, strengthening the patient-to-patient transmission theory. Although none of the identified patients infected with the AIDS virus had sex with Dr. Acer, other sexual partners of his who also were patients could have contaminated the instruments in Dr. Acer's office, Mr. Jaffe said.
To establish a link, though, CDC investigators had to know the order in which patients were treated each day. And, strangely, Dr. Acer's appointment book was missing.
The CDC investigators were able to interview Dr. Acer only once, for about an hour. After that, he went to his lawyer, who blocked all further interviews.
In September, Dr. Acer died.
Soon, Mr. Jaffe's team of investigators discovered a new problem. Dr. Acer was one of the dentists used by Cigna Dental Health of Florida, and as such he would have undergone a routine inspection of his office and practice. But when the CDC asked Cigna Dental Health for their inspection report, it could not be found.
Still another obstacle was finding Dr. Acer's sexual partners. Mr. Jaffe and his team wanted to know if they had the same virus he did and if and when they were treated in his office.
Though Dr. Acer's bisexuality was known, the CDC investigators were unable to identify a single one of his sexual partners. None has come forward voluntarily.
In yet another paper presented in Florence this week, David M. Bell of the CDC calculated that there was only one chance in 42,000 of an infected surgeon's giving the virus to a patient during an operation.
A patient would be two times more likely to die in an airplane crash, or 42 times more likely to die in an automobile crash.
So the question remains: If the odds are so remote, how does the CDC explain the perplexing case of the Florida dentist?
"We can't," Mr. Bell said. "It's a mystery."