Six people were admitted to Flushing Hospital in Queens, N.Y., between Aug. 12 and 23, 1999, with similar enough symptoms of high fever, altered mental status, and headache to make it probable that they were suffering from the same thing.
A routine culture screen for a few suspected bacterial or fungal microbes was negative, although that wasn't surprising since the growing suspicion was that these patients were experiencing an encephalitis-like disease, probably of viral origin. As the illness progressed, it became increasingly apparent that viral encephalitis was the correct diagnosis. Within three weeks of admission, three elderly patients died.
As a first step in the process to identify the cause(s) of illness, the Centers for Disease Control and Prevention in Atlanta sent samples of blood and cerebrospinal fluid collected from each patient, along with brain tissues from the deceased victims, to Dr. Duane Gubler, director of the Division of Vector-borne Diseases Laboratory at the branch of the CDC in Fort Collins, Colo. The answer came back quickly. The illness was probably caused by a single agent, the St. Louis encephalitis virus (SLE). Diagnosis was based on an antibody reaction routinely used in Dr. Gubler's laboratory to identify that encephalitis- causing virus. However, as good and reliable as that test was, it was not specific enough to distinguish among viruses causing encephalitis in the United States and those found elsewhere.
To confound things even more, the Queens serum samples gave a positive but weak reaction, to which the closest match was the St. Louis encephalitis virus. The Fort Collins serologists concluded that it must be SLE, since it was the most likely choice. They might have reasoned that the lack of a strong positive reaction probably related to the specific (new?) strain of SLE involved in the outbreak. A slight difference in protein structure of the viral particles could result in a weakened intensity of the interaction of antibodies with the virus without changing the diagnosis. That they failed to test the sera against the West Nile virus (WNV) was not unusual considering that the WNV agent had never before been encountered in the Western Hemisphere. Therefore, even though the CDC had all the proper reagents and the capability to detect the West Nile virus, the test was never run.
A commonly held adage among pathologists in the United States that has been adopted by many other health professionals as their credo when they encounter the unknown or unfamiliar is: "When you hear hoofbeats think horses, not zebras." In other words, start solving a problem by considering the most likely scenarios first before moving on to more exotic possibilities.
St. Louis encephalitis infections had occurred before in New York state but never in New York City. Finding it there did not seem entirely out of the realm of possibility. Again, the most reasonable explanation for the cause of the encephalitis outbreak seemed to lie in the positive, albeit weak, antibody reaction to the SLE virus. The physicians at Flushing Hospital were notified of the findings, and all personnel involved with the epidemic at the NYC Department of Health and CDC went about the job of determining how many other cases there were throughout the New York area. Since the virus is transmitted to humans by the bite of infected mosquitoes, it was reasoned that it was probably in other places, too. In retrospect, the hoofbeats were from zebras.
By the sixth of September, things began to escalate on all fronts. There were now five confirmed cases, as many as 34 others awaiting final diagnosis, and the first deaths. Although hundreds of native bird deaths had been recorded throughout the city starting in late July, it wasn't until Sept. 9 that exotic birds began dying at the Bronx Zoo. Dr. Tracy McNamara, a pathologist there, quickly became involved in the unfolding drama. After autopsying all the birds that died inside the zoo exhibits, and some that had died outside the zoo, she was no closer to explaining their deaths than before her investigation began. Dr. McNamara suspected, based on the similarity of their presenting pathological conditions, that they had all been killed by the same disease, but she was unable to identify the disease without further testing.
No one working either from the human epidemic or the wildlife epidemic made the connection between the two. More zebras.
A New York State wildlife pathologist examines a dead crow for traces of the West Nile encephalitis virus. He received boxes of dead crows from veterinarians all over the New York metropolitan area.
The next day, another human case was confirmed, and the number of suspected cases rose to 48. This was now more than just a worrisome local health problem in Queens. That evening, specially equipped helicopters hovered above major highways and residential areas deemed most at risk from invasion by mosquitoes and sprayed misty clouds of insecticide (malathion and pyrethriods) in their wake. A general health warning was issued by Dr. Neal Cohen, city health commissioner, about how to avoid contact with both mosquitoes and the insecticides.
By Sept. 13, 11 confirmed cases, three deaths, and a total of 65 probable cases had been identified. Nearly all these were over the age of 50. Numerous reports of new dead bird sightings made the headlines of local and national news, and the infection cast its shadow over a widening range of hosts and territory.
Some mosquitoes collected from Connecticut tested positive. By combining small samples of the same species of adult mosquitoes into "pools" then analyzing each pool for virus, virologists, using the same test as the CDC, showed that Culex pipiens and Aedes vexans harbored the St. Louis virus particles. On Sept. 20, the CDC took another long look at all their test results and began to suspect that the illness was not St. Louis encephalitis after all. The same day, the CDC laboratories in Fort Collins received additional specimens from the scene of the outbreak: tissue samples from dead birds (crows and exotics from the Bronx Zoo). The CDC conducted additional tests on the brain tissues of deceased victims to look for the virus. Dr. Sharif Zaki and his team employed an antibody test designed to detect the common protein shared by all flavivirus members. Their results were positive, revealing the presence of the virus family member protein in the brain tissue sections. This important finding encouraged investigators to proceed to attempt to isolate and A study the virus directly from these human tissue samples.
While all this activity was going forward, on Sept. 21, Dr. Ian Lipkin, director of the Emerging Infections Laboratory of the University of California at Irvine [and now a Columbia faculty member], was invited to join in the hunt by Drs. Leo Grady and Cinnia Huang of the Griffin Laboratory for Virological Studies at the New York State Department of Health in Albany. Grady and Huang submitted five brain samples from autopsies to Dr. Lipkin's laboratory for further analysis.
Dr. Lipkin's team began the work of sequencing the viral RNA genome. Remarkably, by the next day, they had identified a portion of the genome of the flavivirus in three of the brain samples as being related to the West Nile virus, and on Sept. 23, they had extracted the complete length of the virus's single stranded RNA molecule and cloned it. Lipkins's preliminary results were immediately emailed to a special internet site, ProMED (Program for Monitoring Emerging Diseases), a "chat room" for the exchange of preliminary data and ideas regarding epidemics occurring anywhere in the world. It is avidly visited on a daily basis by most scientists who specialize in infectious diseases.
Sequencing the cloned molecule, now converted to a more usable form (DNA) began that day and ultimately led to the identification of thespecific virus responsible for the health problems in Queens. The work was finished in just a little over 24 hours.
Dr. Vincent Deubel of the Pasteur Institute in Paris, reading of Dr. Lipkin's findings on Pro MED, immediately made available sequences of cloned genomes collected from a variety of West Nile virus outbreaks in Romania, Egypt, Israel, Italy, and South Africa, among others. The genomic sequence that Dr. Lipkin's group had determined from the virus in the brain samples from the Queens victims matched exactly with the strain that caused an epidemic in Israel in 1998 -- and with no other. Dr. Lipkin informed the New York State health department and Dr. Steven Ostroff, the CDC epidemiology officer in charge of the epidemic, that his group had determined the complete viral genome sequence. The California team got it right, correctly identifying the causative agent of the encephalitis epidemic as none other than the West Nile virus.
Hurricane Floyd passed through on Sept. 16,breaking the weather conditions that favored the spread of the virus to humans. The epidemic was waning. No new cases were reported after Oct.16. In all, 62 clinical cases and seven deaths had occurred.
Article courtesy of The Journal of the College of Physicians and Surgeons of Columbia University, Winter 2002