False Negatives Complicate Control of Marburg Outbreak
Recombinomics Commentary
April 12, 2005
>> "Nobody really has a sense of where or when it started," said Dr. Thomas Grein, a medical officer in the World Health Organization. "The widespread belief that it began in October is speculation." ........
.
When strange deaths first began to appear in October, mystified local health officials shipped samples of tissue and blood from four children to the United States.
In November, the Centers for Disease Control and Prevention tested them for at least three different types of hemorrhagic fever, including Marburg.
The results, which nearly all agree were accurate, came back negative. But in the tumult of deadly diseases and other health issues that plague this continent, it remains possible that Marburg was present in Uíge even then.
By the end of December, at least 95 children were dead, local health workers say.
How many deaths were Marburg-related is unknown, but even by the grim standards of the continent, it was an alarming number of deaths.
"In October, November, December, we were seeing so many children dying - just children," said Dr. Gakoula Kissantou, 31, the hospital's acting administrator. "It was becoming scarier."
He recalled the doctor in charge of the pediatric ward at the time, Dr. Maria Bonino of Italy, called a meeting with the staff and asked, "What is going wrong here in the hospital?" She herself died in March, a victim of the virus.
It was not until early March that the provincial health officials alerted a W.H.O. representative that they had found 39 suspected cases of Marburg. W.H.O. officials identified 60 possible cases. Angolan authorities then shipped more samples to the C.D.C. in Atlanta. On March 18, 9 of 12 came back positive for Marburg, which by then was claiming more victims by the day. <<
The description of the start of the Marburg epidemic in Angola again places false negatives at the center of an epidemic that is markedly more severe than WHO and media reports would suggest. There are many possible reasons, but the most glaring, which has also been cited in the monitoring of avian influenza in Vietnam, is sample collection and testing.
The samples may be collected under less than ideal situations, packed up, shipped long distances, and then tested. Frequently this process produces false negative results.
The above description indicates there were 95 dead children by the end of December and the first 4 samples tested for Marburg in November were negative. It remains unclear how many of the 95 deaths in the last three months of 2004 were Marburg infections. The latest data by the Red Cross only shows 14 Marburg deaths for those three months. However, even these conservative numbers show 50 more deaths in January and February, yet new samples were not tested until March, after health care workers began to die. By the time Marburg was announced, the number of official dead had risen to 95 for the entire breakout.
However, even the figure of 95 seems to be a lower limit, because only 14 of the 95 deaths in 2004 are classified as Marburg cases. It would appear that the requirement for being an official case is set at a very high bar, and many cases, in part due to collection and testing procedures, slide under that bar. The extreme conditions imposed by the biohazard suits, infection control, and reluctance to get close to patients, will create even more false negatives, which may further limit contact tracing.
Thus, like avian influenza in southeast Asia, setting a high bar for classification of cases allows many to go unnoticed, spreading the virus and creating a situation where the virus becomes endemic to the area.
Now the Marburg virus is spreading in densely populated areas, such as the slums of Luanda, and effective control is becoming increasingly difficult.
Media link