ARTICLE IN BRIEF
- ✓ Debate over differing interpretations of data for guidelines on antibiotic use for post-Lyme syndrome has prompted an investigation by Connecticut Attorney General Richard Blumenthal.
Connecticut Attorney General Richard Blumenthal is investigating treatment guidelines for nervous system Lyme disease issued in October 2006 by the Infectious Diseases Society of America (IDSA). As a part of that investigation, the Attorney General recently subpoenaed documents related to the AAN's Lyme disease guidelines first published May 23 online in advance of the print edition of Neurology (2007;69:91–102).
Both sets of guidelines support conventionally recommended antibiotics for treatment of the infectious disease caused by the tick-borne spirochete Borrelia burgdorferi. The guidelines found evidence of no beneficial effect from using antibiotics for more than 30 days in patients diagnosed with post-Lyme syndrome (PLS). Symptoms include musculoskeletal pain, fatigue, and neuropsychiatric difficulties, without laboratory or clinical evidence of active CNS infection.
The AAN is not a target of the investigation. Rather, the Connecticut Attorney General subpoenaed the Academy for information on how its guidelines for Lyme disease were developed and its relationship, if any, with the IDSA, said AAN General Counsel Murray G. Sagveen.
The investigation specifically seeks to determine whether the effect of the IDSA guidelines is “constraining choices by patients or doctors in a way that could be anticompetitive,” Attorney General Blumenthal told Neurology Today. “If the effect of the guidelines is to deny medical coverage for other kinds of treatment, there might be a violation of antitrust law.”
The investigation, launched last November, is also looking into possible conflicts of interest among the members of the panel that developed the IDSA guidelines, Blumenthal said.
Lyme disease was first identified in the Connecticut city, Lyme, in 1977. The state has the highest incidence of Lyme disease.
Blumenthal launched the investigation at the request of Lyme disease patients and physicians who believe in the efficacy of long-term use of intravenous antibiotics for the treatment of PLS. The International Lyme and Associated Diseases Society (ILADS) has its own treatment guidelines, which recommend extended antibiotic use for chronic Lyme disease.
A RESTRAINT OF TRADE
AAN General Counsel Sagsveen said he has submitted the requested information about the AAN guidelines. But, he added, he is baffled by the idea that the IDSA guidelines restrain physicians from making their own choices — and that the IDSA guidelines are the target of an antitrust investigation. “If the IDSA guidelines are a restraint of trade, then by logical extension isn't any set of medical guidelines a restraint of trade?” he asked.
Henry Allen, a Holland & Knight partner who teaches law at Northwestern University and serves as counsel for the AAN on the Blumenthal investigation, said that the antitrust laws prohibit restraints of trade that are unreasonable in the sense they injure consumers. He noted that “non-coercive clinical guidelines do not even amount to a restraint, much less one that might be deemed unreasonable.”
Moreover, both the Federal Trade Commission and the Supreme Court have recognized that guidelines for clinical patient care are reasonable because they have major consumer advantages, Allen said.
“The biggest advantage is that providing information to consumers saves physicians, on behalf of patients, the trouble of doing their own investigation,” Allen said. “Insurance companies, being consumers too, are also advantaged.”
At the same time, he added, “those who are opposed to these voluntary guidelines have an opportunity to express their ideas as well. The remedy for any mistake is, as one antitrust tribunal put it, not antitrust litigation but more speech — the marketplace of ideas.”
THE AAN GUIDELINES: ON PLS
John J. Halperin, MD, lead author of the AAN treatment guidelines for Lyme disease, a member of the IDSA treatment guidelines panel, and a neurologist at Atlantic Health and Overlook Hospital in Summit, NJ, pointed out that the AAN guidelines' recommendation against long-term treatment of these patients is based on the only three randomized, double-blind, placebo-controlled published trials of antibiotic therapy. Two were published in the New England Journal of Medicine (NEJM) in 2001 (345:85–92). The third trial was published in Neurology in 2003 (60:1923–1930). The subjects had previously received initial courses of antibiotics for acute Lyme disease, but they had ongoing symptoms indicative of PLS.
In the NEJM studies, 129 patients were randomized to receive twelve weeks of antibiotic treatment. At six months, there was no significant difference in the general health and mental health assessments of patients who received placebo versus those who received antibiotics. About 40 percent of patients in each group (placebo or antibiotic) improved in the general health assessment, while 30 percent were unchanged and 30 percent worsened. Several mental health measurements improved in both placebo- and antibiotic-treated patients.
The 2003 Neurology study examined changes in fatigue, cognitive function, and CSF clearance of B. burgdorferi antigen in 55 patients with PLS. At six months, there was significant improvement in fatigue for the patients treated with antibiotics compared to the placebo. The study authors noted that enough patients receiving antibiotic treatment correctly guessed they were in that group to conclude that this finding might be due to “unblinding” and a placebo effect. There was no improvement in cognitive function or clearance of the CSF antigen. The authors also pointed out that the risks of antibiotic therapy do not justify its use as a treatment for fatigue.
Dr. Halperin said he was aware of two additional randomized, controlled, double-blinded studies addressing this issue, one published since the AAN guidelines were issued and the other about to be published. A study in the European Journal of Clinical Microbiology Infectious Disease (2007; 26(8):571–581) indicated no benefit to prolonged antibiotic therapy. The other, which was presented in abstract form at the 10th International Conference on Lyme Borreliosis and Other Tick-Borne Diseases (2005), also indicated no lasting benefit to prolonged therapy.
“I can't in good conscience recommend long-term antibiotics as being of benefit,” Dr. Halperin said.
DEBATE ABOUT CRITERIA FOR GUIDELINES
In large part, the debate about the guidelines boils down to a difference of opinion about criteria for evidence-based medicine. Daniel Cameron, MD, MPH, the upcoming president of ILADS and the chief author of the ILADS Lyme treatment guidelines, charged that the IDSA and AAN guidelines don't take into account evidence-based medicine. He described that as a treatment approach that allows physicians to take into account “their own values, clinical expertise, and patient values” in addition to published research from level 1 studies.
Dr. Cameron contends that the AAN and IDSA guidelines relied on a total patient sample size of only 184 from the three level 1 — randomized, controlled, double-blind — studies. He said the ILADS guidelines relied on the 79- to 92-percent success rate for more than 1,000 patients described in level 2 studies. He added that those level 2 studies are well-designed non-placebo, randomized trials and case series.
But Dr. Halperin told Neurology Today that about half of the references cited in the ILADS guidelines would not typically be included in AAN guidelines. These included presentations at meetings, publications in non-Medline listed journals, review articles, and other material that either has not undergone a full and detailed peer-review process, or includes no meaningful new data. These studies would be graded class 4 by the AAN criteria, Dr. Halperin said.
In fact, he added, when such material is eliminated, the ILADS guidelines includes only three papers (54 patients) not included in the AAN or IDSA guidelines. One, he said, which studied successful treatment of 48 patients with the acute Lyme rash, does not appear relevant to the care of patients with chronic symptoms. The other two are small case series involving two and four patients, which would also be graded as Class 4 evidence.
Dr. Halperin stressed the importance of relying on level 1 studies, in which patients and physicians are blinded for both groups — the experimental drug and placebo. “When dealing with a subjective symptom complex, such as post-Lyme syndrome, blinding and placebo control are particularly important since there are no independent objective measures of disease except how the patient feels,” Dr. Halperin said. “If the patient believes in the treatment, there's a high probability he or she will report improvement. If the doctor believes in it, the doctor will similarly ask questions until some evidence of treatment response is elicited.”
In another difference of opinion about the guidelines, Dr. Cameron said the IDSA and AAN parameters overemphasize the risks of antibiotic therapy. “There really hasn't been any comprehensive harm versus benefit assessment published on chronic Lyme that we can study,” he said. “The Lyme disease patients in the NEJM studies were sicker than the average type 2 diabetic or a patient recovering from a heart attack. [The IDSA and AAN] guidelines mention adverse events, but they don't mention the harm of leaving a chronic Lyme disease patient sick.”
The Neurology study reported that 12 of the 28 patients receiving antibiotics developed diarrhea, two developed minor allergic reactions, and one developed anaphylaxis, which required hospitalization. The NEJM study found that 16 of 64 patients on antibiotic treatments developed adverse affects. Two of the affected patients suffered life-threatening difficulties, one with a pulmonary embolism and the other with a fever and gastrointestinal bleeding.
Dr. Cameron noted that the ILADS guidelines criticized the two studies [cited by the AAN guidelines and published in the NEJM] for “enrolling patients with refractory Lyme disease who were sick for a mean of 4.7 years, despite an average of three courses of antibiotics and for relying only on one treatment protocol.”
“Applying these findings to target populations with characteristics that differ from those included in the trials is inappropriate and may limit options for chronic Lyme disease patients who might benefit from antibiotic treatment,” he said.
The AAN and IDSA guideline panels concluded that given that all studies of this population demonstrated no benefit of prolonged treatment, then any side effects would be too many, Dr. Halperin said. “Given that even in the highly controlled environment of clinical trials, one patient in three had an adverse event and one in 23 had a potentially life-threatening event, it was impossible to justify prolonged regimens,” he said.
Both the IDSA and the AAN Guideline panels reviewed all available peer reviewed literature, including many studies that were not class 1, Dr. Halperin said. “When it came to PLS, there were multiple class 1 studies, all reaching the identical conclusion,” he said. “In this circumstance, particularly in the absence of any studies beyond anecdotal reports contradicting them, the conclusion appears incontrovertible — there is no evidence to support long-term antibiotic therapy.”
Dr. Cameron said physicians should be able to decide for themselves which side they should take in the “differences of opinion” found between IDSA and ILADS. “They're not allowing physicians to use their own clinical judgment and take into account their own values,” he said.
The AAN and IDSA guidelines both, however, include disclaimers that the guidelines are voluntary and they leave ultimate decisions about patient care to the individual physicians. The AAN guideline states the guideline “is not intended to include all possible proper methods of care for a particular neurologic problem or all legitimate criteria for choosing to use a specific procedure. Neither is it intended to exclude any reasonable alternative methodologies.”
“The AAN disclaimer reminds the treating physician that they are responsible for determining the best care for the patient,” AAN General Counsel Murray Sagsveen said.