Joseph J. Burrascano

Lyme disease: point/counterpoint

Lyme disease represents a growing public health threat. The controversial science and politics of Lyme disease have created barriers to reliable diagnosis and effective treatment of this protean illness. Two major clinical hurdles are the absence of a therapeutic end point in treating Borrelia burgdorferi, the spirochetal agent of Lyme disease, and the presence of tickborne coinfections with organisms such as Babesia, Anaplasma, Ehrlichia and Bartonella that may complicate the course of the disease. From a pathophysiologic standpoint, the affinity of Borrelia burgdorferi for multiple cell types and the presence of nonreplicating forms of the Lyme disease spirochete have contributed to persistent infection and failure of simple antibiotic regimens. Newer approaches to the treatment of Lyme disease should take into account its clinical complexity in coinfected patients and the possible need for prolonged combination therapy in patients with persistent symptoms of this potentially debilitating illness. The optimal antibiotic regimen for chronic Lyme disease remains to be determined.

Lyme disease: the quest for magic bullets.

Lyme disease represents a growing public health threat. Recent molecular and genetic studies have confirmed that Borrelia burgdorferi, the spirochetal agent of Lyme disease, is one of the most complex bacteria known to man. Affinity for multiple cell types and the presence of non-replicating forms of B. burgdorferi have contributed to persistent infection and failure of simple antibiotic regimens. The controversial clinical science of Lyme disease has impeded reliable diagnosis and effective treatment of this protean illness. Two major clinical hurdles are the absence of a therapeutic endpoint in treatingLyme disease and the presence of tick-borne coinfections that may complicate the course of the illness. New strategies for the diagnosis, treatment and prevention of Lyme disease are urgently needed.

Treatment of Early Lyme Disease

TO THE EDITOR: Wormser and colleagues (1) reported the results of very short-term (10-day) versus short-term (20-day) antibiotic treatment for patients with early Lyme disease who presented with an erythema migrans rash. Using an on-study analysis, the authors claimed that 84% to 90% of patients had a complete response to one or the other treatment after 30 months of follow-up. These results represent creative mismanagement of the study data (2). Although the study enrolled 180 patients in 3 treatment groups, only 99 patients were evaluable after 30 months of observation, yielding a dropout-plus-exclusion rate of 45% (25% of patients were excluded, and 20% dropped out). Since almost half of the patients were not included in the final analysis of this observational trial, the on-study results are virtually meaningless because the uncounted participants must be considered potential treatment failures (2). Furthermore, an exclusion rate of 25% invalidates the study randomization, and a dropout rate of 20% invalidates the overall study results (2). In the more appropriate intention-to-treat analysis, which we present in the Table, the least stringent response rates (complete response plus partial response) ranged from 49% to 62%, while the most stringent response rates (complete response only) ranged from 44% to 53% at 30 months of follow-up. These results are a far cry from the response rates trumpeted by the authors, and they indicate potential failure of both the short-term and very short-term regimens in a significant number of patients. Table. Intention-To-Treat Analysis of Short-Term Antibiotic Therapy for Lyme Disease An illustrative problem with the analysis was the exclusion of the 5% to 10% of patients who developed a recurrent erythema migrans rash. Although the authors excluded these patients because of the possibility of a new spirochetal infection, a more likely explanation is that the patients had recurrent rashes because of failure of their initial treatment and persistent Lyme disease (3). Thus, manipulation of the study results turned an intention-to-treat failure into an on-study success, and this outcome highlights the problematic data interpretation embraced by the authors. Furthermore, by conservative estimate, at least 41% of patients with early Lyme disease never develop an erythema migrans rash (4). Thus, the study by Wormser and colleagues included only patients whose conditions were easiest to diagnose, making the poor treatment results of short-term therapy even more disappointing. Over the past 3 years, undertreatment of Lyme disease has become institutionalized in the United States. This unfortunate trend has evolved on the heels of a highly flawed study of chronic Lyme disease therapy (5) and publication of manipulated data from early Lyme disease treatment, as seen in the report by Wormser and colleagues and elsewhere (6). It is time to start using well-designed studies and more appropriate statistics in the analysis of Lyme disease therapy in order to assess the gravity and risk of this protean illness. In contrast to the current national trend, we need to examine longer courses of antibiotics to treat persistent spirochetal infection and obtain better clinical outcomes for patients with Lyme disease (6).