Volume 12, Number 5—May 2006
Enrofloxacin in Poultry and Human Health
To the Editor: Following logic similar to that recently used by the US Food and Drug Administration to withdraw approval for enrofloxacin, a recent letter estimated that fluoroquinolone use in poultry could compromise responses to antimicrobial drugs in >24,000 persons per year in the United States (1). However, >99.9% of this estimated risk appears to result from incorrect assumptions. Potentially important corrections include the following: 1) not attributing resistance from foreign travel and human ciprofloxacin use to domestic use of enrofloxacin in poultry (this could reduce the estimated risk by ≈1/3) (2); 2) updating the estimated fraction of human foodborne Campylobacter infections caused by poultry to reflect declines in microbial loads on chicken carcasses since 1992 reduces the estimated risk by a factor of perhaps 1/10 (3) (the cited 90% estimate by Hurd et al.  was intended for use as part of a conservative upper-bounding analysis, not as a realistic point estimate); 3) replacing an assumption that 10% of infected persons would benefit from antimicrobial drug therapy with a more data-based value of 0.6% (4) would reduce the estimated risk by a factor of 0.6/10 = 0.06; 4) replacing an assumption that fluoroquinolones are prescribed for all affected patients receiving antimicrobial drug treatment (rather than, for example, erythromycin) by a more realistic value of fluoroquinolones being prescribed for perhaps ≈50% of patients (2) reduces the estimated risk by a factor of ≈50%; 5) replacing an assumption that all such cases lead to compromised responses with a more data-driven estimate that perhaps ≈17% of patients have compromised responses would reduce the estimated risk by a factor of 1/6 (5); and 6) recognizing that reducing enrofloxacin use may not decrease fluoroquinolone resistance in all Campylobacter spp. from food animals (effect not quantified) (6). Together, such changes reduce the estimated risk by a factor of at least (1/3) × (1/10) × (0.6/10) × (1/2) × (1/6) = 0.00017, or by >99.9%.
More notably, the calculation in (1) also wrongly assumes that the fraction of patients with fluoroquinolone-resistant infections times the fraction of infections caused by poultry gives the fraction of patients with compromised response caused by fluoroquinolone use in poultry. As a simple counterexample, suppose that 80% of all infections were caused by poultry, with the rest caused by something else (e.g., water), and that all and only the 20% of infections caused by the latter source are resistant. Then the procedure in (1) would estimate (80% of infections caused by poultry) × (20% of infections resistant) = 16% as the fraction of resistant infections caused by poultry, even though the correct answer is zero. Thus, the basic logic of the calculation is flawed.
- Collignon P. Fluoroquinolone use in food animals [letter]. Emerg Infect Dis. 2005;11:1789–90.
- Cox LA Jr. Quantitative health risk analysis methods: modeling the human health impacts of antibiotics used in food animals. New York: Springer; 2005.
- Stern NJ, Robach MC. Enumeration of Campylobacter spp. in broiler feces and in corresponding processed carcasses. J Food Prot. 2003;66:1557–63.
- Busby JC, Roberts CT, Lin J, MacDonald JM. Bacterial foodborne disease: medical costs and productivity losses. 1996 [cited 2006 Mar 22]. Washington: US Dept of Agriculture, Economic Research Service. Agricultural economics report 741. Available from http://www.ers.usda.gov/publications/aer741/
- Sanders JW, Isenbarger DW, Walz SE, Pang LW, Scott DA, Tamminga C, An observational clinic-based study of diarrheal illness in deployed United States military personnel in Thailand: presentation and outcome of Campylobacter infection. Am J Trop Med Hyg. 2002;67:533–8.
- DANMAP—Use of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from food animals, foods and humans in Denmark. 2005 Jul [cited 2006 Mar 22]. Available from http://www.dfvf.dk/Files/Filer/Zoonosecentret/Publikationer/Danmap/Danmap_2004.pdf