Animal & Veterinary
In GAO Response, DHHS Cites Studies Showing Human Health Effects Linked to Antimicrobial Resistance
FDA Veterinarian Newsletter May/June 2004 Volume XIX, No 3
In response to a General Accounting Office (GAO) report that said some researchers see little public health risk from resistant bacteria in food-producing animals, the Department of Health and Human Services (DHHS) provided information on 11 studies to show the strong link between resistance and increased risk to human health.
In May, GAO issued its report, “Antibiotic Resistance: Federal Agencies Need to Better Focus Efforts to Address Risk to Humans from Antibiotic Use in Animals.”
The report described the efforts of Federal agencies to address the risks of resistant bacteria resulting from the use of antimicrobials in food-producing animals. The report said that Federal agencies had expanded their efforts in this area, but mentioned that, although many studies have shown the link between the use of antimicrobials in animals and resistant bacteria causing health problems for people, “other researchers contend that the clinical consequences of the transference, if it occurs, is small.”
DHHS developed a response to the points made in the report, and GAO published the response as an appendix in the report.
In its response, DHHS said the report was thorough and generally accurate, but GAO did not consider several reports that establish the risk to human health from certain antimicrobial uses in animals.
The reports that DHHS described focused on the public health risks caused by resistant Salmonella and Campylobacter, which can prolong the duration of illness, and increase rates of bacteremia, hospitalization and death.
The studies also showed that the majority of Salmonella and Campylobacter infections in developed countries are due to antimicrobial use in food animals, DHHS said.
In one study DHHS cited, a researcher studied Salmonella outbreaks that had been investigated by the Centers for Disease Control and Prevention (CDC) between 1971 and 1983 “and found a higher case fatality rate for patients infected with antimicrobial-resistant Salmonella (4.2 percent) than for those with antimicrobial-sensitive infections (0.2 percent),” DHHS said.
In a similar study done in 1987, “among community outbreaks in which hospitalization rates were reported, 57 percent of cases in resistant salmonellosis outbreaks were hospitalized, compared with 24.5 percent in outbreaks caused by susceptible strains,” DHHS said.
“A more recent CDC study of 24 Salmonella outbreaks that occurred between 1984 and 2002 also found that outbreaks caused by resistant Salmonella resulted in higher hospitalization rates than outbreaks caused by susceptible Salmonella,” the DHHS comments said.
“Studies of salmonellosis cases not limited to outbreaks have also demonstrated that resistance is associated with higher morbidity and mortality. In a prospective CDC study of 758 salmonellosis cases, patients with resistant infections were significantly more likely be hospitalized than were those with susceptible infections, even after accounting for underlying illness and prior antimicrobial exposure using multivariate techniques. Patients with resistant infections also tended to be ill longer (median, 10 versus 8 days) and hospitalized longer (median, 5 versus 4 days) than patients with susceptible infections,” the comments said.
Recent studies that have used epidemiological or statistical methodologies to account for factors such as serotype and age that could confound the outcome “have provided further support for the association between resistance in Salmonella and increased morbidity and mortality,” DHHS said. In one case, a researcher after studying Salmonella cases in the United States between 1996 and 2000 found that “antimicrobial resistance was associated with increased hospitalization and bloodstream infections. Patients with Salmonella isolates resistant to any antimicrobial agent or to commonly used agents (cephalosporins, quinolones, or aminoglycosides) were hospitalized more often than patients with pansusceptible isolates, even after controlling for age, race, surveillance site, serotype, and bloodstream infection in a multivariate analysis.”
A large study in Denmark determined mortality rates associated with different drug resistance patterns in S. Typhimurium. “Patients with pansusceptible strains of S. Typhimurium were 2.3 times more likely to die within two years than the general Danish population, whereas patients infected with R-type ACSSuT (resistance to ampicillin, chloramphenicol, streptomycin, sulfonamides, and tetracycline) were 4.8 times more likely to die,” DHHS reported.
Resistance to nalidixic acid, which often leads to increased resistance to fluoroquinolones, can result in higher mortality. The Danish study found that “patients infected with nalidixic-acid-resistant strains were 10.3 times more likely to die than the general population, (and) those infected with strains resistant to nalidixic acid as well as (resistant to) ampicillin, chloramphenicol, streptomycin, sulfonamides, and tetracycline (ACSSuT) were 13.1 times more likely to die,” DHHS said.
In another recently completed Danish study of patients with culture-confirmed S. Typhimurium between 1995 and 2000, researchers found that “patients with nalidixic acid-resistant infections were more likely to have bloodstream infections or die in the 90 days following specimen collection than those with susceptible infections.”
A Canadian study in 1999 and 2000 looked at the increased burden of illness in patients with S. Typhimurium and both definitive phage type 104 (DT104) and antimicrobial resistance. “In this study, after controlling for significant risk factors and confounding variables, including age, hospitalization was 2.3 times more likely to occur among patients whose infections were resistant to at least ampicillin, kanamycin and/or chloramphenicol, streptomycin, sulfamethoxazole, and tetracycline (R-type AK/CSSuT), compared with AK/CSSuT-susceptible patients (p=0.003) and 3.6 times more likely to occur among patients with non-DT104 R-type AKSSuT infections compared with patients with non-DT104 R-type AKSSuT-susceptible infections (p=0.005),” DHHS said.
Studies of Campylobacter show similar problems with resistance. “Several Campylobacter case-control studies in the United States and Denmark have demonstrated a relationship between quinolone resistance and prolonged duration of illness,” DHHS said. Although the GAO report mentioned a Campylobacter study in Minnesota, still “there are several others that GAO ignores,” the comments said. In one 1996-1997 study in Denmark, the researchers found that patients infected with ciprofloxacin-resistant strains of Campylobacter who were treated with fluoroquinolones or other antibiotics were ill for a median duration of 14 days. Patients with susceptible strains recovered in nine days.
In a multi-State case-control study of sporadic Campylobacter cases in the United States in 1998 and 1999, researchers found that resistant Campylobacter could be more virulent than susceptible Campylobacter. When comparing patients who did not take an antidiarrheal medication, those with ciprofloxacin-resistant infections had a nine-day mean duration time for the symptom of diarrhea, compared with a seven-day mean for those with ciprofloxacin-susceptible infections. Patients who did not take an antidiarrheal or an antibiotic suffered the illness for a mean time duration of 12 days if infected with ciprofloxacin-resistant Campylobacter, compared with a mean time of six days for those patients infected with ciprofloxacin-susceptible Campylobacter, which suggests that resistant Campylobacter is more virulent.
In another Danish study DHHS mentioned, researchers evaluated the relationship between resistance in Campylobacter and increases in both bacteremia and mortality. Among patients with culture-confirmed campylobacteriosis from 1995 to 2000, those with fluoroquinolone-resistant or erythromycin-resistant Campylobacter infections “were more likely to have a bloodstream infection or die in the 90 days following specimen collection than those with susceptible infections,” the report said.
The DHHS comments also mentioned the Food and Drug Administration’s proposal to withdraw approval of an enrofloxacin approved for use in poultry water. In his initial ruling, the Administrative Law Judge presiding in the case concluded, “The preponderance of the evidence establishes that fluoroquinolone-resistant Campylobacter results in an increased severity of campylobacteriosis in humans,” the DHHS comments said.
Drug use data
DHHS agreed with the GAO that drug use data are necessary for obtaining a true picture of the extent of resistance and to get a clear idea of the mitigation steps needed to control it.
Also, data from actual use when combined with surveillance of resistance can show stakeholders the extent of the problem.
While GAO recommended that DHHS work with the U.S. Department of Agriculture to develop a plan to collect information about antimicrobial use in food-producing animals, DHHS said “the most useful and reliable data are those maintained by the drug sponsors.” The problem is, though, that the sponsors do not have to present that information to FDA. “Sponsors typically provide a quantity for each of the dosage forms marketed, but the information is not differentiated by animal species, label indication(s), route of administration or geographic region,” DHHS said.
As a solution, the DHHS comments said, the data collection requirements could be changed so the sponsors would present usable data. DHHS added, “This would require notice and comment rulemaking to revise the current regulation.”