By Tom Wilemon | The Tennessean
A Minnesota Department of Health lab technician packages the cerebrospinal fluid of three confirmed meningitis cases in that state to send to the CDC for testing on Oct. 9, 2012.
A Minnesota Department of Health lab technician packages the cerebrospinal fluid of three confirmed meningitis cases in that state to send to the CDC for testing on Oct. 9, 2012. - Hannah Foslien / File / The Tennessean
More people in Maryland were exposed to medicine linked to the fungal meningitis outbreak than in Tennessee, but illnesses here easily quadrupled those reported from the Chesapeake Bay State.
Discuss the fungal meningitis outbreak on Facebook
Continuing coverage of the fungal meningitis outbreak
An article published in The New England Journal of Medicine on Monday showed wide variations in attack rates, but provided no firm conclusion as to the reason. Tennessee had an attack rate of 10.9 infections per 100 people, compared with just 2.4 for Maryland. The national attack rate was 4.7. Tennessee and Michigan had attack rates more than double the national average.
"These attack rates are most certainly underestimates because this outbreak is certainly not over," said Dr. Rachel M. Smith, an epidemiologist with the U.S. Centers for Disease Control and Prevention, who is the lead author of the article.
However, the death rate turned out to be much less than CDC officials initially feared, she said. As of Dec. 10, it was 6 percent of those who became ill. Officials worried it might go as high as 40 percent based on outcomes from much smaller outbreaks of fungal meningitis traced to contaminated spinal injections. Efforts by the CDC, state health departments and health care facilities to contact patients for early diagnosis and treatment saved lives, Smith said.
Possible reasons given for the variation in state attack rates include differences in the degree of contamination in the medicine shipments, the storage times for the medicine and injection practices at the clinics.
However, Smith said CDC staff do not believe some states are missing infections. More than 99 percent of the 13,534 potentially exposed patients had been contacted by Oct. 19.
Another article about the outbreak published in The New England Journal of Medicine last month drew a correlation between storage times and attack rates. Dr. Marion Kainer, state director of health-care-associated infections for the Tennessee Department of Health, was its lead author.
"We found a strong association between the age of the methylprednisolone vials and the rate of infection in one clinic," the prior article stated. "One possible explanation for this observation is that the level of contamination in the vials may have increased over time, with subsequent high fungal burdens present in older vials."
Picture of outbreak
Monday's article in the medical journal includes authors from the CDC and epidemiologists from Tennessee, Michigan, Indiana, Virginia, Maryland, New Jersey, Florida and North Carolina.
With input from multiple locations and national data compiled by the CDC, it offers the broadest picture of the outbreak thus far. The median age for someone sickened in the outbreak was 64. The youngest meningitis victim was 16, while the oldest was 92.
Women accounted for 61 percent of meningitis cases and 60 percent of all fungal infections. While meningitis has been the most deadly illness associated with the outbreak, patients also have been sickened with injection site infections.
The median incubation period from exposure to illness was 20 days, but some patients didn't get sick until 120 days later.
The peak period for new infections was late September to early October.
What batch of the three recalled lots of methylprednisolone acetate a patient got has been identified as a risk indicator. The one linked to the most infections was twice as likely to make a patient sick as a second lot and more than five times as likely to cause an infection as the third lot.
However, Smith said, the reason for the variation in attack rates in the three lots is unclear. For instance, one of the lots might have had a higher attack rate if it had been more widely used, she said.
The majority of the 20 to 50 new cases being reported each week are injection site infections. These present diagnosis challenges because pain is a symptom that patients already suffer.
Contact Tom Wilemon at 615-726-5961 or firstname.lastname@example.org or follow him on Twitter @TomWilemon.