Symptoms Occurring Following Treatment for Lyme Borreliosis
The symptom complex, in other situations summarized under the rubric 'medically unexplained symptoms' (MUS), typically including combinations of fatigue, sleep disturbance, perceived memory and cognitive difficulty, multifocal pain, and other highly disruptive symptoms, causes significant impairment in up to 2% of the general population at any given time, with lesser impairments in substantially more.[110,111] These symptoms have been described with some frequency in patients who have been treated for Lyme borreliosis, leading to the construct PTLDS.
Studies of symptoms following actual LNB are reassuring. The above-cited study of the 2067 Danish patients with definite LNB (Table 2, EFNS criteria), found no evidence of long-term adverse effects on functioning. A systematic review of prior controlled LNB treatment trials, stratifying patients as having definite, probable or possible LNB, found residual symptoms were significantly more frequent in patients without CSF findings of definite LNB and found no patients treated for definite LNB experienced persistent fatigue. A study of 58 Norwegian patients with LNB found none developed cognitive symptoms in the 12 months following treatment. These studies, taken together, support the conclusion that such post-treatment symptoms are not related to CNS infection, a conclusion further supported by treatment trials of additional antimicrobials, which have consistently failed to demonstrate any persisting benefit, while incurring significant treatment related morbidity,[116–120] particularly with unconventional, prolonged regimens. Notably, one retrospective uncontrolled study of 139 definite LNB patients found residua in about one-quarter; this included both residua of initial neurological deficits and of subjective symptoms. Prior work from this group suggests objective residua (pareses etc.) figured prominently.
To address the possibility that these symptoms, in isolation, reflect extra-CNS Bbsl infection, it is helpful to estimate their positive predictive value for Lyme borreliosis. Using often cited extreme estimates that up to 23% of Lyme borreliosis patients develop these symptoms after treatment and that the USA has up to 440 000 Lyme borreliosis cases annually, the symptoms' positive predictive value would be just 1.5%—i.e. a Lyme borreliosis diagnosis based on these symptoms alone would be wrong at least 98.5% of the time.
Multiple studies suggest persistent post-treatment symptoms occur at similar rates in Lyme borreliosis and controls.[123–128] Two—totalling almost 4500 patients[129,130]—found no association between such symptoms and IgG Lyme borreliosis seropositivity. Another, comparing 1786 predominantly EM patients to separately collected controls found persistent symptoms to be somewhat more common in patients—a finding potentially explained by patient expectations, as a large, randomized treatment trial found the best predictors of recovery without late symptoms were initial expectation of recovery and belief that treatment was with antimicrobials, not placebo. As emphasized in the updated US guidelines the comparable rates of this symptom complex in Lyme borreliosis patients and controls raises 'the possibility that this phenomenon, in whole or in part, may represent anchoring to a recent diagnosis of Lyme disease.' In sum, the data regarding the likelihood of persistent symptoms in extra-CNS disease suggest a frequency comparable to controls, although a slight increase might be possible. However, the absence of an association with LNB appears clear.
Recently updated US, German and French clinical guidelines all recommend against additional antimicrobials in such patients, highlighting risks of prolonged treatment—risks including not only the complications and costs of prolonged administration of unnecessary antimicrobials but also the risk of delay in correct diagnosis in patients who are seeking help with disabling symptoms—symptoms which may have been incorrectly attributed to ongoing or, worse, 'chronic' Lyme disease.
Brain. 2022;145(8):2635-2647. © 2022 Oxford University Press