OUP user menu

Letter of response to Stricker et al.

Beth White, R.A. Seaton, T.J. Evans
DOI: http://dx.doi.org/10.1093/qjmed/hcs228 202-203 First published online: 6 December 2012

ARTICLE

ARTICLE

ARTICLE

Sir,

We appreciate Dr Stricker et al.’s interest in our article describing the use of outpatient parenteral antibiotic therapy in the treatment of Lyme disease.1 They highlight the finding that, of the 47 patients for whom intravenous antibiotic therapy was recommended, the 13 that received prolonged antibiotics appeared to do better than those who received the currently recommended duration of therapy (14–28 days) and then provide a statistical analysis to demonstrate that this difference was statistically significant. Of these 13 patients, 7 received 2–4 weeks of parenteral ceftriaxone followed by a median of 3 weeks of oral doxycycline (range 1–6 weeks), 5 obtaining a definite improvement in symptoms; 3 received prolonged parenteral therapy (each receiving a 6-week course of ceftriaxone), 2 with a definite improvement in symptoms. Three actually had to stop intravenous antibiotics early due to side effects from the treatment and so went on to receive a subsequent course of prolonged (2–5 weeks) oral antibiotics. Our study was not designed or powered to determine the outcome of different lengths of antibiotic treatment in groups of patients with different disease presentations. Subgroup analysis carried out in this fashion is fraught with errors in interpretation. We feel strongly that it is not possible to infer a result of statistical significance from a small heterogenous subgroup of patients from an observational cohort study.

Dr Stricker et al.2 also follow this analysis by stating that our findings support that of a recently published study from his group which they state demonstrated a significant improvement in cognitive function with 6–12 months of intravenous ceftriaxone in patients with chronic neuroborreliosis. First, in keeping with our comment above we do not feel our article provides evidence to either support or discourage prolonged antibiotics in this group of patients. Second, we would urge caution in accepting Stricker et al.’s study findings as evidence to support prolonged antibiotic therapy; there was no control group used, patients were not randomized in their treatment duration and their measure of treatment success was based on a change in patient-assessed symptom severity, focussing on four subjective symptoms (myalgia, arthralgia, fatigue and ‘cognition’ or ‘brain fog’). In addition, the authors have not considered or described adverse treatment events associated with prolonged antimicrobial therapy and indeed the reasons for some of the cohort receiving different treatment durations.

In our opinion, therefore, neither our observational study nor that of Dr Stricker et al.2 provides robust evidence to support more prolonged therapy in this group of patients. If there are concerns that the four randomized controlled trials3–5 whose conclusions discourage the use of prolonged antibiotic therapy are not robust, then the onus is on these clinicians to develop a larger, well-designed randomized controlled trial with robust outcome measures to provide clarity on this issue.

References

View Abstract