Chronic Lyme drugs cause life-threatening skin detachment

“Fake Lyme, real Lyell! When non-compliance with HAS recommendations leads a patient to intensive care… “

Doctors from the Reference Center for Bullous Diseases at CHU Henri-Mondor and from the Reference Center for Tick-borne Diseases at CHI Villeneuve-Saint-Georges report the case of a serious drug eruption that occurred in a patient who was treated unjustifiably with a combination of anti-infectives for suspected chronic Lyme. They deliver their analysis in this column, published in full on lequotidiendumedecin.fr.

CONTRIBUTION – In June 2018, the French National Authority for Health (Haute Autorité de Santé, or HAS) published recommendations for good practice in the management of Lyme borreliosis or Lyme disease. In particular, these recommendations define the symptoms/persistent syndrome after a possible tick bite (SPPT) and propose a standardized treatment based on specialized regional hospital centers that have established close relationships with local physicians. The aim is to respond to a major public health challenge, i.e. to fight against medical wandering, to reduce treatment delays and, above all, to avoid inappropriate drug prescriptions for patients who turn to the healthcare system for poorly explained polymorphic symptoms.

This article was written in response to the recent treatment in our hospital (reference center for serious drug eruptions) of a patient in her forties (the patient’s consent was obtained before publication of this article) for Lyell syndrome (or toxic epidermal necrolysis), a serious drug eruption characterised by extensive epidermal and mucous membrane detachment that can lead to death [1].

A cocktail of six anti-infective drugs

This woman, a school teacher, whose main history was an erythema migrans treated with doxycycline seven years previously (according to the recommendations in force at the time) and recurrent genital herpes, had for eight months presented mechanical joint pain, episodes of asthenia and ill-defined food intolerances. The diagnosis of late disseminated Lyme disease had been suspected and a broad combination of anti-infectives was prescribed approximately three weeks prior to admission, combining praziquantel [anti-worm], ivermectin [anti-parasite], cotrimoxazole [an antibiotic also known as sulfamethoxazole/trimethoprim or Bactrim], fluconazole [an antifungal also known as Diflucan], rovamycin [an antibiotic and antiparasitic also known as Spiramycin], desloratadine [an antihistamine also known as Clarinex], as well as homeopathy.

Two weeks after the start of this treatment, the patient developed oral erosions with a febrile maculopapular exanthema. Extensive bullae led to the diagnosis of Lyell syndrome 48 hours later. Intrinsic accountability (timing) was used for cotrimoxazole, fluconazole and rovamycin, but extrinsic accountability (drug notoriety) tended to favour cotrimoxazole, a drug at high risk for Lyell syndrome [2,3].

Fifteen days of resuscitation for negative Lyme serology

The patient was hospitalized in dermatology and quickly transferred to intensive care for respiratory failure. The maximum skin detachment was 35% and was associated with extensive ocular, laryngeal, tracheobronchial and esophageal involvement. After 15 days of resuscitation and a total hospital stay of one month, the patient returned home with epidermal skin and healing mucosal erosions. Lyme disease serology (ELISA test) was negative for IgG and IgM.

One quarter of preventable drug eruptions

Stevens-Johnson Syndrome (SJS) and Lyell Syndrome are drug-induced in 85% of cases: they are the most serious form of drug eruptions. Overall mortality in the acute phase is 15%, but can be as high as 40%. More than half of the patients suffer long-term ocular, cutaneous, and psychological sequelae, which have a significant impact on quality of life [4]. The majority of SJS/Lyell are induced by high-risk drugs, including cotrimoxazole [3]. However, a French study has shown that nearly 25% of cases are preventable because they are linked to inappropriate use of medication: inappropriate indication, relapse despite a history of allergy, or self-medication. Allopurinol and cotrimoxazole are at the top of the list of inappropriate indications [5]. For our patient, it was indeed the prescription of several drugs, including at least one considered high risk, for a suspicion of undocumented Borrelia infection, which occurred after seven years of symptom-free interval since the episode of erythema migrans, and not complying with the HAS recommendations, which led to the occurrence of Lyell syndrome.

Three-part test for Lyme disease

The diagnosis of active Lyme borreliosis is based on a three-part test:

– A clear exposure to tick bites: here proven since there is a history of erythema migrans,

– A compatible clinical presentation: here unlikely in the face of non-specific clinical signs and multiple joint pain of mechanical schedule, unlike Lyme joint borreliosis where in more than 90% of cases it is a monoarthritis. Moreover, the clinical signs occurred after several years of symptom-free interval after treatment of migrant erythema,

– A positive serology in ELISA confirmed by Western-Blot: here the serology was negative, which is very rare in late disseminated forms, apart from severely immunocompromised patients. It should be noted that even if the serology had been positive in our patient, as the clinical signs were not very suggestive, this diagnosis of Lyme borreliosis would have also remained uncertain. A positive serology is a marker of Borrelia encounter and antibody production, but not a marker of disease activity. The clinical signs are at the forefront.

A trial antibiotic course for 28 days

Furthermore, the treatment is based on national recommendations (HAS, 2018; and Scholarly Associations Consensus Conference, 2019 [6]), similar to those in Europe and internationally. The diagnosis of Lyme borreliosis is sometimes difficult, since it can mimic many other pathologies. This is why in case of strong diagnostic suspicion, despite an incomplete three-part test, the HAS recommendations recommend the use of a doxycycline antibiotic test for a maximum of 28 days. These recommendations nevertheless emphasize the following points:

• Any test antibiotic therapy exceeding 28 days should only be prescribed in research protocols validated by a personal protection ethics committee.
• If there is no improvement in symptoms at 3 weeks, the antibiotic therapy should not be renewed, given the risk of side effects estimated to be greater than the hypothetical benefits expected for the patient.

To date, no acquired resistance of Borrelia to antibiotics has been described and five randomized studies have shown the lack of superiority of antibiotic therapy lasting more than one month, the lack of benefit of drug combinations, as well as the potentially serious side effects to which patients could be exposed in case of prolonged antibiotic therapy.

Doxycycline, the most effective oral molecule

The two most effective molecules and therefore the most used for the treatment of Lyme borreliosis are doxycycline, orally, and ceftriaxone, intravenously. It should be noted that the patient did not receive either of these two molecules in her multiple combination therapy. Praziquantel and ivermectin are ineffective antiparasitic agents for Borrelia. Fluconazole is an antifungal agent ineffective on Borrelia, probably prescribed to avoid fungal infections secondary to the anti-infectives. Rovamycin is moderately effective on Borrelia and is a second-line treatment for severe allergies. Azithromycin, also of the macrolide family, is preferred, but with fewer side effects. Finally, cotrimoxazole is not recommended for Borrelia infections because it is not very effective and should only be used for documented infections and in the absence of other alternatives given the potentially serious side effects (see Vidal).

Finally, since ticks can transmit other pathogens (Anaplasma, Rickettsia, etc.), the hypothesis of co-infections (rare with Borrelia) or other tick-borne vector-borne diseases is routinely raised in the management of these patients, but here again, the clinical signs are telling, with often obvious biological signs. In case of trial antibiotic treatment, doxycycline is also the first choice.

Benefit/risk balance

In infectious pathology, as in medicine more generally, the treatment proposed to the patient must be based on a favorable balance between the expected benefits and the potential risks. While risks are fortunately rare but often unpredictable for many drugs, benefits must be established conscientiously and honestly with valid and reproducible scientific data, and must be constantly re-evaluated and updated in light of new data. This is the foundation of modern scientific, ideally evidence-based medicine. Scholarly societies and other regional, national and international medical institutions thus issue, following a consensus methodology, opinions based on defined levels of evidence with varying strengths of recommendations [7].

Failure to comply with these recommendations primarily involves patients, who are both the losers and the victims of a game of deception in which they are most often unknowingly confronted and without knowing the rules, as illustrated by this clinical case. Doubts about the best treatment to propose must always benefit the patient and be based on scientific data. The rule of “primum non nocere” (first do no harm) prevails. Therapeutic risks must be discussed honestly with the patient. The unknown must remain in the field of medical research, which places this uncertainty in a regulatory framework that is structurally the most protective of patients [8].

It is through the promotion of research, the maintenance of the scientific honesty of each practitioner in the face of up-to-date medical knowledge, and respect for the trust of patients that is placed in us at each consultation that this type of event can be avoided.

References

[1]      Duong TA, et al. Severe cutaneous adverse reactions to drugs. Lancet 2017;390:1996–2011.

[2]      Arimone Y, et al. Updating the French method for the causality assessment of adverse drug reactions. Therapie 2013;68:69–76.

[3]      Mockenhaupt M, et al. Stevens-Johnson syndrome and toxic epidermal necrolysis: assessment of medication risks with emphasis on recently marketed drugs. The EuroSCAR-study. J Invest Dermatol 2008;128:35–44.

[4]      Ingen-Housz-Oro S, et al. Health-related quality of life and long-term sequelae in survivors of epidermal necrolysis: an observational study of 57 patients. Br J Dermatol 2020;182:916–26.

[5]      Chaby G, et al. Severe cutaneous adverse reactions due to inappropriate medication use. Br J Dermatol 2018;179:329–36.

[6]      Gocko X, et al. Lyme borreliosis and other tick-borne diseases. Guidelines from the French scientific societies. Med Mal Infect 2019;49:296–317. (LymeScience discussion)

[7]      Brozek JL, et al. GRADE Guidelines 30: The GRADE Approach to Assessing the Certainty of Modelled Evidence – an Overview in the Context of Health Decision-making. J Clin Epidemiol 2020.

[8]      Harris IA, Naylor JM. Double standards in clinical practice ethics. Med J Aust 2014;200:76. 

Authors

Dr. Nicolas de Prost, réanimateur, AP-HP, Henri-Mondor Hospital, Reference Center for Toxic Bullous Skin Diseases and Serious Drug Eruptions TOXIBUL, Créteil

Dr. Alice Raffetin, infectious disease specialist, Reference Center for Tick-borne Diseases, Ile-de-France/Hauts-de-France, Villeneuve-Saint-Georges Hospital Center, Villeneuve Saint-Georges

Dr. Sébastien Gallien, infectious disease specialist, Reference Center for Tick-borne Vector-borne Diseases, Ile-de-France/Hauts-de-France, AP-HP, Henri Mondor Hospital, Créteil

Dr. Saskia Ingen-Housz-Oro, dermatologist, Reference Center for Toxic Bullous Skin Diseases and Serious Drug Eruptions TOXIBUL, AP-HP, Henri Mondor Hospital, Créteil