Current issues of ACP Journal Club are published in Annals of Internal Medicine


Antimicrobial prophylaxis for Lyme disease was not indicated after a deer-tick bite

ACP J Club. 1993 May-June;118:81. doi:10.7326/ACPJC-1993-118-3-081

Source Citation

Shapiro ED, Gerber MA, Holabird NB, et al. A controlled trial of antimicrobial prophylaxis for Lyme disease after deer-tick bites. N Engl J Med. 1992 Dec 17:327:1769-73.



To assess the risk for infection with Borrelia burgdorferi and the effectiveness of prophylactic antimicrobial treatment in reducing this risk in patients with a recognized deer-tick bite.


Randomized, double-blind, placebo-controlled trial with at least 12 months follow-up.


An area of southeastern Connecticut where Lyme disease is endemic.


387 people (mean age 27 y, 51% adults; 57% women; 15 people were enrolled twice) who had been bitten by a deer tick within the previous 72 hours. Exclusion criteria were a bite by a tick of a different species, allergy to penicillins, pregnancy, or previous antimicrobial treatment. 93% were followed for 1 year.


205 participants received amoxicillin, 250 mg, and 182 received a placebo (either a capsule with lactose or, for children, a liquid suspension without the antibiotic) 3 times a day for 10 days.

Main outcome measures

Symptomatic or asymptomatic infection with B. burgdorferi. Symptomatic infection was defined as the presence of either erythema migrans at the site of the bite or symptoms of either early disseminated Lyme disease (e.g., Bell palsy) or late Lyme disease (e.g., arthritis) with seroconversion. An asymptomatic infection was defined by the occurrence of seroconversion without signs or symptoms of Lyme disease. Serum was obtained at entry and 6 weeks and 3 months later. Participants were interviewed when serum was collected and by telephone 7 days, 6 months, and 1 year after enrollment.

Main results

53 of the 344 ticks (15%) analyzed with polymerase chain reaction were infected with B. burgdorferi. Symptomatic infection developed in 2 participants, both receiving placebo. There were no asymptomatic seroconversions and no late manifestations of Lyme disease. Infection with B. burgdorferi developed in 1.2% of placebo group patients compared with 0% of amoxicillin group patients (P = 0.22). The low risk for infection in the placebo group makes the estimate of the efficacy of amoxicillin imprecise.


The risk for infection with B. burgdorferi after a recognized deer-tick bite was so low that prophylactic antimicrobial treatment was not indicated.

Sources of funding: In part, National Institutes of Health and Wyeth-Ayerst Laboratories (medication).

For article reprint: Dr. E.D. Shapiro, Department of Pediatrics, Yale University School of Medicine, 333 Cedar Street, New Haven, CT 06510-8064, USA. FAX 203-785-7194.


The article by Shapiro and colleagues addresses the issue of appropriate management of patients bitten by ticks that are potential vectors of Lyme disease. The study was done in southeastern Connecticut, an area endemic for disease, and was methodologically rigorous. The risk for infection with B. burgdorferi after a recognized deer-tick bite was estimated at 1.2% (95% CI 0.1% to 4.1%), a risk that is at the lower end of several published estimates, but the confidence interval includes the range of probabilities reported in other studies. Because of the low risk for infection, this study did not have the statistical power to show a protective efficacy of empiric antibiotic therapy. A total of 1400 persons would need to be randomized to detect a statistically significant 3-fold reduction in risk for early infection. Although not formally analyzed, when the results of the few available randomized trials are pooled, there does seem to be a statistically significant benefit of empiric therapy, and infection only occurred in subjects who received placebo. Because some sequelae are too infrequent, randomized trials are not well suited to evaluate the efficacy of empiric antibiotics for the prevention of the late complications of Lyme disease. A cost-effectiveness analysis reported a possible benefit of empiric antibiotic therapy in terms of lower overall costs and late complications of Lyme disease if the risk for infection after a tick bite exceeds 3.6% (1), a probability exceeded in only a few areas of the United States. Issues surrounding repeated courses of antibiotic therapy, antibiotic resistance, and use of medical services by treated individuals have not been adequately addressed. Until additional information is available, routine antimicrobial therapy after a recognized tick bite in most areas is probably not recommended. Some physicians, however, prescribe empiric therapy for pregnant women and, because the risk for infection increases with feeding duration, for patients who present with highly engorged ticks.

Brian S. Schwartz, MD, MS
Johns Hopkins School of Hygiene and Public HealthBaltimore, Maryland, USA


1. Magid DJ, Schwartz BS, Craft J, Schwartz JS. Prevention of Lyme disease after tick bite: a cost-effectiveness analysis. N Engl J Med. 1992;27:534-42.