The report by Shadick and colleagues [1] confirms data from previous studies and strengthens the concerns of Lyme disease support groups that many patients who receive apparently adequate antibiotic treatment for Lyme disease do not recover completely [1, 2]. We recently described a large cohort of patients (n = 215) who had contracted Lyme disease in the mid-1980s in Westchester County, New York [3]. Although all patients in our larger sample were Lyme-antibody positive, the similarity between our findings and those of Shadick and colleagues are striking. Patients in our study had a relapse rate of 28% and a reinfection rate of 18% compared with 26% and 21%, respectively, in the cohort of Shadick and colleagues. We also found a high prevalence of persistent symptoms that were related either to previous articular or neurologic involvement or to the presence of diffuse arthralgia and fatigue. As the authors point out and as we have described, this post-Lyme disease syndrome shares many features with the chronic fatigue syndrome and fibromyalgia [4]. Psychometric testing showed a decrease in verbal memory scores in some of these patients [5].

An interesting finding in our study and in Shadick and colleagues' study was that patients whose initial antibiotic treatment was delayed were more likely to have a poorer outcome. Further, Shadick and colleagues had a disproportionately large number of patients who presented without erythema migrans and who had presumably developed later stages of Lyme disease before treatment, another correlate of incomplete recovery [3]. This finding suggests that persistent infection, residual injury, or other ongoing processes (possibly immunologic) may be important in the pathogenesis of these symptoms. Because Lyme disease is being recognized and treated earlier in areas where it has long been endemic, the incidence of incomplete recovery in the 1990s should be lower than that reported in these studies.