Tuesday, Nov. 4, 2004, is of course presidential Election Day. Political pollsters are already busy scoping public-opinion tea leaves as to predictions for a second term.

A less momentous but significant public health prediction has been announced for Tuesday, July 1, 2004. Epidemiologist Marc Lipsitch, Harvard University School of Public Health, Boston, couched his foreboding forecast.

"We predict that by July 1, 2004 - in the absence of a vaccine - 41 percent of pneumococci will be dually resistant, with 5 percent resistant to penicillin only and 5 percent to erythromycin only. By that date," he said, "we predict that in Georgia, 65 percent of the vaccine-type pneumococci will be dually resistant and that in Minnesota, 67 percent of the strains included in the vaccine will be dually resistant."

Georgia and Minnesota, he explained, were two sites for which long-term serotype data were available.

Lipsitch is senior author of an article in the current Nature Medicine, released online March 3, 2003. Its title: "Geographic diversity and temporal trends of antimicrobial resistance in Streptococcus pneumoniae in the United States."

"There were two findings in our paper," Lipsitch observed. "The first is that we analyzed the force of the geographic variation among different parts of the U.S. The prevalence of the bug's resistance to two of the major drug classes, in S. pneumoniae and the variation, were much greater than would be expected from chance. And we were also able to reject another explanation, which is that individual drug resistance strains had managed to do well in a few regions, but not in other regions. By rejecting those two explanations, the only other interpretation that we could come up with is that probably different levels of antibiotic use in certain regions have different levels of resistance.

"Our second finding," Lipsitch continued, "was that we used a mathematical model to analyze existing trends and project future trends in S. pneumoniae. We found that resistance to both classes of drugs we looked at - erythromycin macrolides and penicillin beta-lactams - was increasing. In particular, strains resistant to both antibiotics simultaneously were increasing much faster than strains resistant to either one individually. So we made a series of projections outward to the middle of 2004, and the baseline prediction was that if trends we studied from 1996 to 1999 continue, we would expect about 41 percent of strains to be resistant to both classes of antibiotics. And another 5 percent or so to be resistant to each of those, but not the other; so a total of 51 percent resistant to one or more of those two antibiotics."

Drug Resistance Vs. Sensitivity And Vice Versa

Lipsitch pointed out: "One factor has changed significantly since our study period began in the mid-90s. It's the introduction of a vaccine that will change levels of resistance, so we made some preliminary predictions, trying to take that into account. The S. pneumoniae vaccine should reduce resistance because it targets a certain set of serotypes - antigenic molecules - that happen to be associated with drug resistance."

Turning from the immunology to actual use of the antibiotics, he opined: "It documents for the U.S. that probably antibiotic use is driving significant differences in what proportion of different strains people are exposed to, just based on residence. Which means that we really have the impact of these selective forces showing itself in the U.S., which therefore gives greater support to the need to control unnecessary antibiotic use.

"It also projects that the problem is really getting worse quickly, and that the rise of multiresistance means that use of either class of antibiotic risks selecting not only resistance for itself, but for the other ones, if they tend to be in the same bacterial strains."

The Nature Medicine paper leads off by introducing S. pneumoniae as "a leading cause of bacteremia [sepsis], sinusitis, otitis media [middle-ear infection], bacterial meningitis and pneumonia worldwide." These blood and brain infections, Lipsitch noted, make S. pneumoniae a "truly life-threatening bacterium."

He and his co-authors reported collecting population-based data through active surveillance of invasive pneumococcal infections at eight sites around the U.S., covering some 21 million people. They tested these collected isolates for antimicrobial susceptibility, as opposed to antibiotic resistance. In 1998, the proportion of penicillin resistance varied from 14.7 percent in the state of New York to 35.1 percent in Tennessee.

"One of my areas of ongoing work in the laboratory," Lipsitch told BioWorld Today, "is assessing the effect of these resistance genes on the fitness of the microorganism. A lot of people have advocated reducing antibiotic use - as I would - in the hopes that not only would we reduce the rate of increase of resistance but might even reverse it, allowing the sensitive strains to come back.

"For that to happen, if we stopped all antibiotic use today and there was no difference between resistant and sensitive strains, then we would expect the current prevalence to stay fixed - if there's no selection for resistance and no selection against it."

In Low-Drug Europe, No Patient Goes Untreated

"It's really the human abuse of antibiotics that's driving bacterial resistance," Lipsitch suggested. "In one too-common scenario, it's the patient coming to the M.D. and between them they decide on an antibiotic prescription. That's what the patient asks for and sometimes not. There's a need for better education of both groups and efforts to minimize unnecessary use.

"The fact that this is possible is clear from the contrasts in Europe," he pointed out. "If you compare different countries in Europe, there's about a fourfold downward difference in the use of beta lactam (penicillin) antibiotics per patient per year. Yet, in none of those countries are people going untreated for life-threatening infections. It's not that health givers are rationing these drugs and putting peoples' health at risk. So that suggests to me that there's really an opportunity to use antibiotics in a limited, wise way - and also the possibility of overusing them quite dramatically. The U.S.," he concluded, "is somewhere in the middle of that range."