The recent report by Shadick and associates [1] provides valuable data on the long-term symptoms that may be found in patients with a history of Lyme borreliosis [1]. In understanding these data, however, it would be helpful to clarify the total number of symptomatic patients in both the Lyme disease and control categories. The statement that "most patients with Lyme disease in our study were without symptoms at the time of clinical assessment" [1] appears to be contradicted by the information presented in the authors' Table 3, which indicates that 61% of patients had frequent arthralgias, 26% had unusual fatigue, and 47% had trouble sleeping. It appears that of the 25 patients judged not to be persistently impaired, many were excluded because of the lack of a positive serologic test result or inadequate earlier classic manifestations, even though they met the previously stated criteria for Lyme borreliosis.

Clarification of the antibiotic responsiveness of the patients would also be helpful. It appears that 5 of 10 patients retreated with antibiotics improved in response to such treatment. One of the nonresponders, patient 12 in Table 6, had autopsy confirmation of central nervous system spirochetes. This information would make one suspect that continued infection, rather than fixed damage from past infection, may explain persistence of symptoms in at least some patients. The investigators do not state which antibiotic regimens were used to achieve response in the five responders. Knowledge of the clinical characteristics of the antibiotic responders compared with those of the nonresponders would be interesting.

Major clinical questions confronting the clinician caring for patients with Lyme disease are whether the persistently symptomatic patients are still infected, and, if so, how they can be helped [2]. This study gives us clues that I hope will help to improve the substantial morbidity and occasional mortality associated with this illness.