To summarise the existing data for the treatment of early Lyme disease.

MEDLINE (January 1966 to December 1997) was searched for both articles and reviews using the following keywords: Lyme disease, Lyme, early Lyme disease, Lyme borreliosis, Borrelia burgdorferi, erythema migrans borrelios, erythema migrans, penicillin, amoxicillin, beta-lactam(s), cefuroxime axetil, cefuroxime, cephalosporin(s), doxycycline, tetracycline(s), erythromycin, azithromycin, clarithromycin, macrolides. Further articles were identified from a manual search of the reference lists of retrieved articles and review papers.

Study designs of evaluations included in the review

Randomised, prospective controlled trials. Only trials published in the English language were included in the review. Economic analysis of diagnosis and treatment strategies for early LD ere also included.

Specific interventions included in the review

Antimicrobial treatments that were subdivided into penicillins and tetracyclins (phenoxymethylpenicillin 250mg/900mg/600mg, amoxicillin 500mg with or without probenecid, doxycycline 100mg), cephalosporins (ceftriaxone 1g, cefuroxime axetil 500mg), and macrolides (azithromycin 500mg/ 250mg, roxithromycin 150mg). Other antibiotic therapy in some control groups included Erythromycin 250mg, tetracyclin 250mg and minocycline 100mg.

Participants included in the review

Non-pregnant adult patients with early Lyme disease (LD). Mean patient age ranged from 38-54 years. Trials evaluating therapy of late manifestations of Lyme disease (e.g. neuroborreliosis, recurrent arthritis) or disease prevention were excluded.

Outcomes assessed in the review

The outcome measure of interest was not specified in the inclusion criteria. The specific outcome measures reported by included studies was the resolution of erythema migrans (EM) and symptoms at the end of the treatment period (sometime to resolution), and the (prevention of) development of late LD (which was usually classified as major or minor) as well as the incidence of adverse drug reactions at follow-up. The length of follow-up of included studies, where specified, ranged from 1-24 months.

How were decisions on the relevance of primary studies made?

The authors do not state how the papers were selected for the review, or how many of the authors performed the selection.

There was no reported validity assessment of each individual study other than the collection of data on how many of the randomised participants were included in the final analysis. However, design limitations of some of the included studies, such as small sample size, lack of double-blinding, inadequate measure of compliance and a short follow-up period, were commented upon. The authors do not state how the papers were assessed for validity, or how many of the authors performed the validity assessment.

The following information was obtained from each study: clinical trial design, number of patients randomised versus analysed, criteria for eligibility, regimen of study antibacterials, outcome measures used, length of follow-up period, age of patients, number of patients with disseminated infection at study entry, time from beginning of illness to start of study antibacterials, presence of adverse effects and results at the end of the treatments and follow-up. It was not stated how many reviewers were involved in this process.

How were the studies combined?

Studies were combined in a narrative summary.

How were differences between studies investigated?

The authors do not state how differences between the studies were investigated.

Eleven randomised prospective controlled trials were included in the review with a total of 1,213 patients. Three trials evaluated the use of penicillins and tetracyclins (number of patients randomised=247; number of patients analysed=219), three trials investigated the use of cephalosporins (number of patients randomised=436; number of patients analysed=373) and five studies investigated macrolides (number of patients randomised=530; number of patients analysed=422). Three cost effectiveness analysis were also included.

Evidence from the clinical trials suggest that phenoxymethylpenicillin, amoxicillin, amoxicillin/probenecid, tetracycline, doxycycline, and cafuroxime axetil are equally effective for the treatment of early Lyme disease. However, the available studies do not have sufficient power to detect small but potentially clinically relevant differences in the efficacy between the individual drugs. Clinical failures with erythromycin, roxithromycin and azithromycin indicate that macrolides cannot be recommended as first line agents for this infection.

According to the findings of two trials, cefuroxime seemed to cause more diarrhoea than doxycycline. One of these studies also reported the incidence of the Jarisch-Herxheimer to be higher among cefuroxime treated participants. Both studies found the incidence of photosensitivity to be significantly higher among doxycycline treated participants than those on cefuroxime. One further study reported that phenomethylpenicillin was significantly better tolerated than minocycline. Azithromycin caused significantly more diarrhoea than amoxycillin, and more persistent or severe headache.

Included studies support the use of treatment courses that range from 10 to 21 days.

Three cost-effectiveness analysis that dealt with diagnostic and treatment strategies for early Lyme disease were included in the review. Empirical treatment with oral antibacterials of patients presenting with EM is cost effective when compared with intravenous ceftriaxone therapy. There have been no studies evaluating the incremental cost-effectiveness of various oral antibacterials for the treatment of early Lyme disease so there is no proven economic basis to choose among options. However, intuitively, it makes sense to choose strategies that minimise non compliance as the direct medical costs to treat sequelae appear to be large. For patients with myalgic symptoms alone (and no rash) no further treatment or diagnostic testing is warranted. For patients with both myalgic symptoms and a rash resembling EM as well as a history of a previous tick bite, laboratory testing with the ELISA and confirmation with the Western blot is the most cost-effective strategy. Based on economic modelling, empirical treatment with oral antibacterials of persons who present with a history of tick bite in endemic areas (>1.5% prevalence) may be cost effective.

Oral beta-lactam antibitotics (phenoxymethylpenicillin, amoxicillin, cafuroxime axetil) and oral tetracyclines are effective treatment modalities for early Lyme disease. Oral macrolides are considered second-line agents as their clinical efficacy has been less than that of beta-lactams and tetracyclines.

The review states the objective and includes pre-specified inclusion/exclusion criteria. MEDLINE was the only electronic database that was searched and therefore some important information may have been missed. No attempt was made to look for unpublished studies and the presence of publication bias can not be ruled out. In addition, only English language trials were considered for inclusion in the review, which may cause bias due to the exclusion of RCTs published in any other language. Information about the methodology of the review process (such as how decisions on the relevancy of primary studies were made, whether more than one reviewer conducted data extraction, and how discrepancies were resolved) is very limited. There was no reported structured validity assessment of included studies (such as the use of a validity checklist), however the limitation of the overall methodological quality of the studies was discussed briefly in the review. Study details are presented but without a formal investigation of heterogeneity it is unclear why studies were not pooled.

The authors' conclusions seem to follow from the results, but should be treated with caution owing to the above limitations.

The authors state that further studies are warranted to determine the appropriate duration of therapy for treatment of early Lyme disease.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.