The Infectious Diseases Society of America (IDSA), along with the American Academy of Neurology (AAN), and the American College of Rheumatology (ACR), recently published a public request for comments on the “2019 Draft Guidelines for the Prevention, Diagnosis and Treatment of Lyme Disease.” Below is the official response from Global Lyme Alliance (GLA).
The 100-page guidelines document attempts to cover various aspects of Lyme disease, including prevention, acute Lyme disease, neurological Lyme disease, co-infections, and more. GLA purposefully focused on seven (7) specific areas of the Draft Lyme Disease Guidelines to comment on. The 7 areas are handpicked for two key reasons: they represent the most glaring oversights and omissions in the guidelines and there is strong scientific evidence to refute them.
GLA focused its comments on 1) the overall misses of the guidelines, 2) Lyme disease diagnostics, 3) tick bites and prophylactic treatment, 4) neurological Lyme disease, 5) post-treatment Lyme disease, 6) treatment for persisting Lyme infections, and 7) chronic Lyme disease.
1 – General Comments
General comments on entire draft, pages 2-69, lines 45-1605: Throughout the 2019 draft revised guidelines, an overriding concern is the generation of false positive diagnoses and misattribution of symptoms to Lyme disease. In contrast, there is little, or no concern voiced about the possibility of false negative diagnoses and misattribution of Lyme disease symptoms to other etiologies. Obviously, all patients, whatever their ailment, should be accurately diagnosed in a timely manner. Testing methods more sensitive and reliable than the CDC standard two-tier test (STTT), as concluded by 40 Lyme disease academic and government specialists attending a Banbury Conference at Cold Spring Harbor Laboratory, and a year later by the Tick-Borne Disease Working Group, are desperately needed to differentiate between those who do and those do not have Lyme disease. Individuals suffering from Lyme and other tick-borne diseases (TBDs) who are not promptly diagnosed are likely contributing, at least in part, to the accumulation of patients suffering from post-treatment Lyme disease syndrome (PTLDS). Because of the severe consequences of missed or delayed diagnosis of Lyme disease, it is extremely important to minimize false negative diagnoses as well. Specific guidance on how best to avoid false negative diagnoses would be invaluable to medical care providers.
2 – Lyme Diagnostics
Diagnostic Testing for Lyme disease, page 10, lines 228-229: It states that “IgG seronegativity in a patient with prolonged symptoms (months to years) essentially rules out the diagnosis of Lyme disease…”. This is only true if the patient with prolonged symptoms has no history of other objective, clinical measures of Lyme disease. Also, it is noteworthy that research has shown that seroreactivity to Borrelia burgdorferi antigens varies between individuals and during the course of disease within an individual. Therefore, a static test at one point in a dynamic disease process is a poor diagnostic tool. The IDSA/CDC criteria requiring that five of 10 IgG bands be present for positive diagnosis does not make sense immunologically as it suggests that a patient with only four bands does not have Lyme disease. How does one then explain away the presence of four distinct antibodies highly specific in their ability to recognize B. burgdorferi antigens? Furthermore, work published by researchers at Northeastern University and Johns Hopkins University have shown in vitro and in vivo (in a mouse model of Lyme borreliosis), respectively, that at different stages of growth B. burgdorferi expresses a distinct repertoire of antigens and degrees of antibiotic tolerance. While in vivo studies are ongoing to identify these persister forms in patients, it is premature to dismiss the possibility that persisters expressing variant antigens, and the antibodies they elicit, exist in PTLDS patients. Therefore, the Western immunoblot of the CDC standard two-tier test (STTT) may not include the full repertoire of antigens recognized by patients with prolonged symptoms. Their absence from the STTT could result in late stage patients failing to meet the “five of 10 IgG bands” criterion for a positive diagnosis. As the field transitions from use of the STTT to a modified two-tier test (MTTT) this same concern about whether the ‘correct’ antigens are targeted remains.
Another factor to consider is that continued B. burgdorferi-specific IgM reactivity has been observed in IgG seronegative patients. This scenario is confusing, because it can lead to a false positive Lyme diagnosis. However, there also are IgM-positive/IgG-negative patients who were previously diagnosed with Lyme disease, based upon presentation of erythema migrans or other CDC criteria. The Johns Hopkins University SLICE studies suggest that a diversity serological profiles exist in PTLDS patients; these include those who are IgM-positive/IgG-negative when tested months after cessation of treatment. In an initial case series of 60 patients, while 43% were IgG-positive, 11% were IgM-negative. Among these IgM-negative patients, most were IgG-negative as well. In short, the lack of B. burgdorferi-specific IgG does not necessarily preclude continued disease just as the continued presence of B. burgdorferi-specific IgM and/or IgG is not always indicative of active infection. Acknowledgement of these important, albeit nuanced, facts is absent from the revised guidelines.
3 – Tick Bites and Prophylactic Treatment
Tick bites, prevention, and prophylaxis of Lyme disease – What diagnostic tests should be used following tick bite?, pages 19-20, lines 453-467: It states, “We recommend against testing for B. burgdorferi in an Ixodes tick following a bite”. While true that the presence of a pathogen does not reliably predict the likelihood of clinical infection, there still is valuable information to be gained by testing ticks. Primary care physicians, even in Lyme-endemic areas, are not always familiar with the tick species common to a given location or the pathogen(s) they may carry. The stated rationale for the recommendation not to test is that “Even in areas that are highly endemic for Lyme disease, patients presenting with an Ixodes tick bite have a low probability of developing Lyme disease…”. This statement is not referenced and is unsupported by the fact that in highly endemic areas, in the northeastern U.S. in particular, the carriage rate for B. burgdorferi can be 50% or greater and such high carriage rates correlate with a higher incidence of Lyme disease. Tick removal and sending it for testing is a relatively rare event and thus should not contribute significantly to unnecessary antibiotic prescriptions. A major benefit to tick testing is that if no pathogens are found, this information would put the patient’s mind at ease. If instead, the tick tests positive for one or more pathogens, this will focus the physician’s and patient’s attention on what symptoms to look for and better inform a treatment strategy if such symptoms arise.
The guidelines also suggest that “Anticipatory guidance is recommended so that a prompt diagnosis of Lyme disease (as well as other Ixodes tick transmitted infections) can be made should a patient develop symptoms”. However, given the inadequate familiarity most physicians have regarding ticks and tick-borne agents in their area, let alone in an area(s) to which a patient may have traveled and been bitten, there is no basis for thinking such anticipatory guidance can be provided. It is hard to rationalize why the guidelines would advocate for physicians and patients to make decisions having less rather than more information in hand, especially when the benefits of a prophylactic single dose of Doxycycline are substantial and the risks negligible.
4 – Neurological Lyme Disease
Neurological Lyme disease – For which neurological presentations should patients be tested for Lyme disease?, page 36, lines 840-841: It states that “In patients with cognitive decline the guidelines recommend against routine testing for Lyme disease.” Global Lyme Alliance vehemently disagrees with this statement because it does not take into consideration the extensive evidence in peer-reviewed medical journals describing the cognitive decline experienced by Lyme neuroborreliosis patients. Such studies include patients with PTLDS, as defined by the IDSA-proposed case definition, who experience cognitive decline as well as PTLDS patients with neuropsychiatric symptoms linked to neuroimmune responses. This latter reference also argues against the guidelines’ claim that “No studies suggest a convincing causal association between Lyme disease and any specific psychiatric conditions”. If patients present with otherwise unattributable cognitive decline, and they live in a Lyme-endemic area, why should a physician not consider Lyme disease in their differential diagnosis? To not do so risks a missed/delayed diagnosis, and it is well-established that the earlier the treatment for Lyme disease is initiated the more positive the prognosis for recovery and cure. It also is recognized that cognitive symptoms can vary depending on a patient’s age, so a “one size fits all” statement that patients with (otherwise) unexplained cognitive decline should not be tested for Lyme disease seems at odds with a careful process of differential diagnosis and best-practice medical care.
5 – Post-treatment Lyme Disease
Prolonged symptoms following treatment of Lyme disease, page 61, lines 1412-1419: Reference is made to studies of “patients appropriately diagnosed and treated for Lyme disease” who describe either “persisting or recurrent fatigue, musculoskeletal pain, neurocognitive and other non-specific subjective symptoms”. These are in fact patients clinically defined by Rebman et al. as having PTLDS and yet reference to this seminal study and specific mention of this PTLDS patient population is missing and should be included in the body and bibliography of the guidelines. In this same section it states that long-term “symptoms appear to subside over time…”. For patients suffering from PTLDS, most continue to have debilitating symptoms. In fact, a Dutch study found an average of 1.7 disability-adjusted life years lost due to persisting symptoms attributable to Lyme, and even longer for some patients.
6 – Treatment for Persisting Lyme Infections
Prolonged symptoms following treatment of Lyme disease – Should patients with persistent symptoms following standard treatment of Lyme disease receive additional antibiotics?, pages 62-63, lines 1445-1449 and 1464-1468: Four randomized, controlled trials are cited as evidence against repeated antibiotic treatment (references #317, 321, 319). A careful statistical analysis of the trials and their design calls into question the conclusion that repeated antibiotic treatment has no benefit. DeLong et al. showed that two trials conducted by Klempner (reference #317) had sample sizes too small to detect true minimum clinically-important differences, particularly in SF-36 scores. The guidelines state that in study by Fallon et al. (reference #321) “A cognitive index score at week 24 did not differ between treatment and control groups.” While true that objective measures such as cognitive test results did not show differences between PTLDS patients and controls, it is important to note that PTLDS patients in the study had to invest considerably more effort to achieve the same test scores. This would argue against the notion “that this phenomenon, in whole or in part, represents anchoring bias…”. Additionally, the authors did note a positive effect on fatigue, similar to that observed in the Krupp trial (reference #319), and Fallon and co-authors highlighted the need for further study. In reviewing these clinical studies, DeLong et al. concluded that “primary outcomes originally reported as statistically insignificant were likely underpowered.” At the very least, DeLong’s analysis should be cited in the IDSA guidelines as evidence that further study of repeated antibiotic use should be very carefully designed, executed, and interpreted.
Regarding continued or repeated antibiotic treatment for persistent symptoms, the guidelines state that “A body of literature conducted in animal models has raised hypotheses of microbiological persistence”. While correct, this statement is wholly disingenuous as it fails to reference any of the studies and, worse, neglects to acknowledge that experimental results are presented that support the hypotheses put forward. It also is stated that “Moreover, animal models cannot reproduce the human experiences of fatigue and pain, and it is unlikely that any animal study can give reliable insight into the biology of humans experiencing such symptoms following treatment of Lyme disease.” Again, this statement neglects a significant body of literature describing animal models of human fatigue and chronic pain. Such studies have been instrumental in advancing our understanding of the pathogenesis of these human conditions and have spurred the development of novel treatments. This statement also fundamentally misrepresents the intent of the diverse animal studies conducted that provide evidence of persistence of spirochetes post-antibiotic treatment of infected animals. The purpose of using animal models as articulated by Monica Embers, Ph.D. (Division of Bacteriology and Parasitology, Tulane National Primate Research Center, Tulane University Health Sciences Center) is “to understand the etiology of post-treatment Lyme disease syndrome (PTLDS), namely whether or not the spirochetes persist post-treatment and could thus contribute to chronic symptoms.” Symptoms experienced by PTLDS patients are not restricted to fatigue and pain, so one cannot reasonably discount the results of animal findings supporting spirochetal persistence because they are purported not to recapitulate all the symptoms experienced by humans.
Recent human evidence that continued bacterial presence may contribute to long-term symptoms was published by Jutras et al. in a 2019 PNAS article. They found that B. burgdorferi peptidoglycan, a component of the cell wall, was recovered from 94% of synovial fluid samples from Lyme arthritis patients, with specific IgG responses. Many of these patients had previously been treated with antibiotics. As peptidoglycan is shed from actively growing live bacteria, it’s at least plausible and worth further inquiry to study whether bacteria persist in these patients. If not, one is left with a less plausible explanation that foreign bacterial antigens must remain in inflamed tissues for weeks, months, or even years after (presumed) effective antibiotic-mediated killing of B. burgdorferi. With such intriguing findings at least hinting at the possibility of persistent organisms in Lyme disease patients, the authors of the revised guidelines should consider specifically referencing these animal and human studies rather than covering the entire topic by citing one publication (reference #320) and summarily discounting the results presented in so many other peer-reviewed scientific journals.
7 – Chronic Lyme Disease
Prolonged symptoms following treatment of Lyme disease – Chronic Lyme disease, page 64, lines 1474-1479: It states that “The term ‘chronic Lyme disease’ as currently used lacks an accepted definition for either clinical use or scientific study, and it has not been widely accepted by the medical or scientific community”. Although this statement is correct with respect to the lack of an accepted clinical definition, at least 68 publications in peer-reviewed scientific and medical journals spanning 1985 to 2019 describe the chronic infectious and persistent nature of Lyme disease. While treatment of patients with Doxycycline or other standard-of-care antibiotics is quite effective when provided within the first few weeks of infection, no clinical studies have demonstrated complete clearance of spirochetes; just elimination of symptoms for some, but not all patients (i.e., PTLDS patients). When early treatment is ineffective or initial diagnosis is delayed, B. burgdorferi can avoid pharmaceutical and/or immune clearance and spirochetes have an opportunity to disseminate and cause persistent disease. Literally a dozen or more bacterial pathogens are capable of establishing persistent infection and associated chronic disease. It would be truly remarkable if B. burgdorferi were unable to do the same. To our knowledge, there is no scientific or medical evidence to suggest they are incapable. Whether persistent disease is synonymous with persistent infection is an important scientific question worthy of objective consideration and further careful investigation, rather than recrimination, disparagement and dismissal. The revised guidelines would stand on firmer scientific/medical footing were it to acknowledge that the question of persistent disease vs. infection is still an open question, rather than suggesting the latter has no evidentiary support whatsoever.
To conclude, the content and bibliography of the 2019 revised IDSA guidelines fails to acknowledge evidence or reference published scientific and medical studies that could and should convey a more nuanced understanding of the complexities of Lyme disease diagnosis, symptomatology, treatment, and treatment failure. A more inclusive, open-minded, and informed approach to conveying information can only benefit the Lyme disease physician and patient community, as it will better serve to enhance co-operation, reduce controversies that divide the IDSA and ILADS ‘camps’, and ultimately reduce the likelihood of false negative and false positive diagnoses.
Learn more about important GLA-funded peer-reviewed Lyme disease research, ranging from basic science, treatment, to chronic Lyme disease.