Treatment of the disease
Lyme disease is treated with antibiotics and the earlier the treatment is started the better the outcome. However the optimum drug, dose and duration are not known.
See our Guidelines page for other guidelines and guidance referred to within the UK.
Clinical Knowledge Summaries (CKS) for Lyme disease does not go into treatment detail, and for anything but early treatment at the erythema migrans stage recommends referral to a specialist. Unfortunately UK consultants see few Lyme disease cases and experience is spread thinly across the UK’s GPs. Although a House of Lords debate supported a call for development of a national network of interested NHS clinicians, the NHS has no specialists in Lyme disease.
Choice of antibiotic
It is recognised that it is important to have a drug with good tissue penetration, intracellular action and good CSF penetration. The antibiotics normally quoted in guidelines are doxycycline, amoxicillin, cefotaxime or ceftriaxone as these are the drugs that have been used in clinical trials. Because these trials have shown very variable outcomes (38% – 100% successful outcome), and because investigations have shown that the bacteria can survive conventional courses [eg 1] many doctors and researchers are looking for better treatment.
There is an opinion that metronidazole and tinidazole may be necessary against the cyst form of the bacteria . There is also a view that ketolides (such as azithromycin) which show very high effectiveness in the laboratory , may have a place in treatment. Neither of these classes of drug have yet been tested in clinical trials in patients with Lyme disease.
European Guidelines for Lyme Neuroborreliosis (EFNS)  state, under a discussion of effective agents in late Lyme NeuroBorreliosis (LNB) that “there are no randomised treatment studies of European late LNB”.
See the guidelines above for generally recommended doses of each drug. Because of the many treatment failures in clinical trials, it has been suggested that higher doses may be necessary to effectively eradicate the bacteria, but there have been no European trials exploring this.
Because the bacteria can penetrate the spinal fluid it is important that high enough antibiotic concentrations are reached in the CSF. Research has shown that doxycycline at 200mg twice a day achieves the necessary levels in the CSF much faster than with the often recommended 100 mg twice per day . This is likely to be important in short courses of 14-28 days but, because of the long half life of doxycycline, may not be so important in longer courses: evidence is lacking.
Duration of treatment
The EFNS guidelines  state that “There are no comparative controlled studies of treatment length in European late LNB” ( symptom duration > 6 months).
The view of many doctors is that duration of therapy should be guided by clinical response. A study looking at cases in Glasgow found that those treated for longer than the guidelines suggested did better.
Patients might experience a worsening of symptoms on starting treatment. This is termed a Jarisch-Herxheimer Reaction (JHR) after the two people who documented it. CKS  only mentions this with respect to pregnant women, but it applies to all patients.
CKS states “People may mistake this for an allergic reaction and stop their antibiotics. Provided the symptoms are not severe and there is no evidence of an allergic reaction (such as urticaria), they can be advised to continue.”
Note that a European trial found that unlike syphilis, in which the JHR occurs in the first 24 hours of treatment, “We suggest that Jarisch–Herxheimer-like reactions may be prolonged and may occur late during treatment.” 
1) Honegr, K, D et al. 2004. “Long term and repeated electron microscopy and PCR detection of Borrelia burgdorferi sensu lato after an antibiotic treatment.” Central European journal of public health 12 (1) (March): 6-11.
2) Brorson & Brorson 2004. An in vitro study of the susceptibility of mobile and cystic forms of Borrelia burgdorferi to Tinidazole. International Microbiology 7:139–142
3) Hunfeld & Brade 2006 Antimicrobial susceptibility of Borrelia burgdorferi sensu lato: What we know, what we don’t know, and what we need to know. Wien Klin Wochenschr 118/21–22: 659–668
4) Dotevall & Hagberg 1989 Penetration of Doxycycline into Cerebrospinal Fluid in Patients Treated for Suspected Lyme Neuroborreliosis. Antimicrobial Agents and Chemotherapy, July 1989, p. 1078-1080
5) Mygland et al 2010 EFNS guidelines on the diagnosis and management of European Lyme neuroborreliosis. European Journal of Neurology 2010, 17: 8–16
7) Oksi et al 2007 Duration of antibiotic treatment in disseminated Lyme borreliosis. European Journal of Clinical Microbiology and Infectious Diseases 2007, 26 (8): 571-81