Authors: Alex Guri, Keren Mahlab-Guri, Michael Giladi, Michal Yakubovsky
Categories: Case Report, Cat-scratch disease, Fever of unknown origin, Jarisch–Herxheimer reaction, Doxycycline, Case report
Source: IDCases
Authors: Alex Guri, Keren Mahlab-Guri, Michael Giladi, Michal Yakubovsky
The term "Jarisch–Herxheimer reaction" (JHR) describes a self-limited but potentially life-threatening systemic inflammatory response characterized by symptoms such as fever, tachycardia, headache, myalgia, hyperventilation and hypotension following the initiation of effective therapy, most commonly in the treatment of spirochetal infections. JHR typically occurs within 1--2 h but sometimes starts a few hours later and can last from 12--24 h. The reaction was named after Adolf Jarisch and Karl Herxheimer, two researchers, who, between the late 19th and early 20th centuries, reported similar reactions in syphilis patients treated with mercury [1], [2]. Cat scratch disease (CSD) is caused primarily by Bartonella henselae, a fastidious, gram-negative bacillus found worldwide in bacteremic cats [3]. Typical CSD presents in approximately 90 % of cases and is characterized by self-limiting regional lymphadenopathy or lymphadenitis.
Fever of unknown origin (FUO) is a rare but well-recognized manifestation of atypical CSD that occurs more frequently in older patients. Data on the efficacy of antibiotic therapy in patients with CSD-related FUO remain limited [4].
Here, we present the case of a teenage boy with FUO who developed an adverse reaction following the administration of doxycycline and was retrospectively diagnosed with CSD-related JHR. to the best of our knowledge, this is the first case report describing Jarisch–Herxheimer reaction in a patient with cat scratch disease.
A 14-year-old Caucasian male from a rural village in southern Israel presented to the Pediatric Emergency Department with a prolonged, unexplained fever lasting 9 days. He was subsequently hospitalized for further investigation. The patient reported daily fever spikes with shivering accompanied by headaches, weakness, and thirst with polydipsia. He had no respiratory, gastrointestinal, or urinary complaints. His daily hobby was dirt bike riding. He also mentioned having close contact with cats, dogs, horses, cattle, and geese, both on his family's farm and at friends' farms. Additionally, he denied a history of travel, being sexually active, smoking, using recreational drugs, or consuming unpasteurized milk or dairy products.
Approximately two years before admission, the boy was diagnosed with attention deficit hyperactivity disorder and was occasionally prescribed methylphenidate. However, he had not taken the medication in the month before his hospitalization.
On physical examination upon admission, the patient had a fever of 39.1°C (PO), a heart rate of 100 bpm, and a blood pressure of 109/60 mmHg. He was slightly tachypneic (24 breaths per minute) with a peripheral oxygen saturation of 99 % in ambient air. He appeared sick and pale, but no rash or eschar was observed, and there was no lymphadenopathy. A systolic murmur (2/6) was auscultated, but there were no stigmata of endocarditis. The remainder of the physical examination was unremarkable.
The patient was admitted to the pediatric ward for further evaluation. Blood cultures were negative. The laboratory evaluation demonstrated elevated inflammatory markers, mild hyponatremia, and an increased lactate dehydrogenase level, while renal and hepatic function tests were normal; values are detailed in Table 1.Table 1Key abnormal laboratory findings on admission.Table 1ParameterResultReference RangeWhite blood cell count15.8 × 10⁹/L5.5–15.5Neutrophils13.1 × 10⁹/L(83 %)1.8–8.0Lymphocytes1.9 × 10⁹/L(11.9 %)1.2–5.2Hemoglobin10.8 g/dL13–16 g/dLPlatelet422 × 10⁹/L150–450Sodium133 mEq/L136–145Lactate dehydrogenase308 U/L125–220C-reactive protein5.54 mg/dL0.0–0.5Units: L, units per liter; g/dL, grams per deciliter; mEq/L, milliequivalents per liter; mg/dL = milligrams per deciliter
Serological tests for Epstein–Barr virus and cytomegalovirus suggested past infections. The chest X-ray results were normal, and an electrocardiogram revealed sinus tachycardia without other abnormalities. A transthoracic echocardiogram revealed normal systolic and diastolic function, with minimal mitral and tricuspid regurgitation and no valve vegetation.
At this point, the differential diagnosis, considering the clinical presentation of prolonged fever, neutrophilic leukocytosis, mild hyponatremia, and elevated inflammatory markers in an early adolescent male with extensive animal exposure, raises concerns about a zoonotic or vector-borne infectious etiology. Given the absence of rash or eschar but the presence of daily fever spikes, headache, and systemic symptoms, differential considerations include brucellosis (despite the lack of unpasteurized dairy consumption, exposure to cattle remains a risk factor), leptospirosis (considering environmental exposure and polydipsia), and Q fever (Coxiella burnetii), which is endemic in Israel and associated with exposure to livestock and farm animals. Additionally, relapsing fever (Borrelia spp.) and an atypical rickettsial infection (such as Rickettsia typhi, endemic murine typhus or, less likely, Rickettsia conorii, Mediterranean spotted fever) should be considered, particularly given the patient’s frequent outdoor activities and contact with animals known to harbor vectors. The presence of a systolic murmur without vegetation makes acute infective endocarditis less likely to be diagnosed at this point.
Blood cultures were negative, as was a blood smear for tick-borne relapsing fever. Serological tests were ordered for Q fever, brucellosis, B. henselae, Rickettsia conorii, and Rickettsia typhi.
Given the patient's exposure to farm animals and pets and the possible differential diagnosis of Rickettsial infections, he was started empirically on intravenous doxycycline on the second hospitalization day. The patient’s clinical course is summarized in the timeline shown in Fig. 1.Fig. 1Timeline of the patient’s clinical course.Fig. 1
Approximately 60 min after starting the antibiotic infusion, the patient developed severe myalgia and chills, as well as rigor, flushing, and facial erythema. His temperature was 38.2°C (PO), with a heart rate of 110 bpm, a blood pressure of 105/65 mmHg, and a peripheral oxygen saturation of 98 %. Intravenous methylprednisolone was administered. Oral ibuprofen was added due to myalgia. No respiratory symptoms were observed. A diagnosis of JHR was considered, although an allergic drug reaction was the primary differential diagnosis. In addition, the possibility of gram-negative bacteremia and sepsis was considered. Doxycycline was discontinued, and empirical treatment with ceftriaxone and azithromycin was started. Myalgia and erythema resolved gradually during the next 24 h. Fever persisted until the seventh day of hospitalization. A chest and abdominal CT scan were normal, and tests for antinuclear antibodies, rheumatoid factor, complement components C3 and C4, and stool calprotectin were within normal ranges. Serological tests for Q fever, brucellosis and rickettsial infections were negative. Ultimately, B. henselae enzyme immune assay (EIA) serological tests were positive for both IgG and IgM antibodies, each with a titer of 100, corroborating the diagnosis of CSD-FUO syndrome [4], as EIA has been shown to be highly specific for the diagnosis of CSD [5].
CSD-FUO syndrome is a rare atypical manifestation of CSD. In contrast to typical CSD, CSD-FUO is characterized by a prolonged fever lasting approximately four weeks, either continuous or episodic. Lymphadenopathy is present in only 20 % of cases, and other manifestations, which are distinct from those of typical CSD, such as hepatosplenic space-occupying lesions, ocular involvement, and multifocal osteomyelitis, may also occur [4].
Five months after hospitalization, the patient underwent evaluation at an allergy clinic. Allergic reactions to doxycycline were excluded by a negative skin prick test (10 mg/ml), negative intradermal tests (0.0001 mg/ml, 0.001 mg/ml) and negative drug challenge with an oral suspension (10 mg, 90 mg).
Cytokine measurements were not obtained during the acute reaction, because these assays are not routinely performed in our hospital and are not informative in patients with fever and active infectious disease. Therefore, the diagnosis of JHR was supported by the characteristic timing of symptom onset, their self-limited course, and the absence of clinical features typical of immediate drug hypersensitivity. The subsequent complete exclusion of doxycycline allergy during formal allergy evaluation further reinforced that the reaction represented JHR rather than an allergic event.
At the follow-up visit 6 months after hospitalization, the patient reported that he had returned to good health. The final diagnosis of CSD was made, and the reaction following the initial administration of doxycycline was diagnosed as JHR.
With the advent of penicillin as the primary treatment for syphilis in the 1940s, JHR became widely recognized and frequently reported. For example, Mahoney et al. (1943) reported that some syphilis patients experienced general malaise, mild headache, fever, increased penile lesion pain, and regional lymphadenopathy shortly after treatment initiation [6].
The classical explanation of JHR suggests that it results from an inflammatory response triggered by the release of proteins and toxins following the rapid death and destruction of treponemes, making the reaction more common in primary and secondary syphilis due to the higher bacterial load. Subsequent observations have suggested that JHR is an inflammatory process induced by the activation of the cytokine cascade during spirochete degradation [7].
In subsequent years, it has become evident that JHR can manifest across various spirochetal diseases following the initiation of appropriate antibiotic therapy, including infections with Borrelia recurrentis and Borrelia burgdorferi, the agents of louse-borne relapsing fever and Lyme disease, respectively, leptospirosis, and others [7], [8], [9], [10]. In addition to penicillin, JHR has been reported following treatments with tetracyclines and, more recently, antimicrobials such as cephalosporins, carbapenems, quinolones and macrolides [11].
JHRs are predominantly associated with spirochetes; however, similar reactions have rarely been reported in certain nonspirocheal intracellular pathogens, such as Coxiella burnetii (Q fever), Brucella species and Mycobacterium tuberculosis [12], [13], [14], [15].
To the best of our knowledge, while B. henselae is an intracellular pathogen, systemic symptoms characteristic of classic JHR have not been reported in patients with CSD. Two publications reported the occurrence of JHR in association with CSD, but we believe these cases do not represent typical systemic JHR. The first case described a patient with ocular CSD who experienced worsening neuroretinitis following antibiotic treatment. This reaction, however, was limited to a deterioration in visual function and was not accompanied by systemic symptoms or signs [16]. The second report involved a 12-year-old female with hepatosplenic CSD who presented with prolonged fever and abdominal pain. The authors interpreted the lack of clinical improvement after seven days of treatment with doxycycline and rifampicin as indicative of JHR. However, given that fever in patients with hepatosplenic CSD can persist for more than two months and that no new symptoms occur within 48 h of initiating treatment, we question whether this case truly represents JHR [4], [17], [18].
In our case, the medical team, confronted with the dramatic onset of rigors, severe myalgia, facial erythema, and fever shortly after antibiotic administration, initially proposed two main an allergic reaction to doxycycline and sepsis. Accordingly, the patient was treated with corticosteroids and broad-spectrum antibiotics. However, negative blood cultures and the rapid resolution of symptoms—with the exception of fever, which prompted hospitalization—raised the possibility of a JHR.
All the observed symptoms, including facial erythema, have been previously described in the JHR, as noted by Kadam et al. [19]. Additionally, the patient had no history of atopy to support a drug allergy. The diagnosis of JHR is primarily clinical, and in most cases additional testing is unnecessary when the characteristic timing and symptom pattern are present [20]. Although elevations in specific cytokines have been demonstrated during JHR in research settings [21], cytokine assays are not routinely performed and lack diagnostic specificity in patients with fever and active infection. Therefore, differentiation from drug hypersensitivity relies on clinical assessment rather than laboratory biomarkers. Ultimately, the possibility of an allergic reaction, including a doxycycline challenge, was excluded following an allergy workup.
A review of the published literature reveals that no universally accepted definition or standardized diagnostic criteria for JHR exist. Historical reports differ in the required symptom constellation, timing, and degree of systemic involvement [1], [7], [9], [11], [19], [20]. Despite this heterogeneity, the consistently described core features-abrupt onset of fever, chills, rigors, myalgia, and flushing within hours of initiating appropriate antimicrobial therapy-form the clinical basis for recognizing JHR and align closely with the presentation observed in our patient. Given the absence of definitive diagnostic criteria and the inherent limitations in confirming this syndrome through laboratory testing, the strong resemblance of our patient’s clinical course to the characteristic features described in the literature supported the diagnosis of a JHR.
In contrast to typical CSD, which is generally managed in outpatient settings, presents with short-lived fever and is not associated with overt bacteremia, CSD-FUO syndrome can involve prolonged fever lasting several weeks and may be associated with B. henselae bacteremia, as reported by Welch et al. [21]. In this context, the occurrence of JHR, as in our case, is not surprising.
The lack of awareness of JHR in CSD patients, coupled with the low incidence (2 %–4 %) of CSD-FUO syndrome among CSD patients, may partly explain why JHR has not been previously reported in this population [4]. Increased awareness among physicians regarding the possibility of JHR in CSD patients, particularly those with CSD-FUO syndrome, could lead to better identification of such cases. Moreover, distinguishing JHR from an allergic reaction is essential, as mislabeling patients as allergic to doxycycline could unnecessarily deny them the drug of choice for treating conditions with significant morbidity, such as rickettsiosis, brucellosis, Q fever, and plague.
AG and MY were involved in the literature search and drafted the manuscript. KMG provided information on immunological aspects, conducted medical follow-up on the patient and invited him to her clinic to rule out a doxycycline allergy. MG made critical revisions to the manuscript. All text in the manuscript was written by AG, MY and MG. All the authors read and approved the final manuscript.
All authors confirm that they have read and approved the final manuscript entitled “Jarisch–Herxheimer Reaction Following Doxycycline Therapy in an Adolescent with Cat Scratch Disease: A Case Report.”
The work is original, has not been published previously, and is not under consideration elsewhere.
All authors meet the ICMJE criteria for authorship and agree to be accountable for all aspects of the work.
Michael Giladi: Writing – review & editing, Validation, Supervision, Methodology, Investigation, Formal analysis, Data curation. Keren Mahlab-Guri: Writing – review & editing, Validation, Methodology, Data curation, Conceptualization. Michal Yakubovsky: Writing – original draft, Validation, Supervision, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Alex Guri: Writing – original draft, Resources, Methodology, Investigation, Data curation, Conceptualization.
Not applicable.
Written informed consent was obtained from the patients’ parents (guardians) for publication of this case study. A copy of the written consent is available upon request.
ChatGPT used only for language correction and phrasing refinement, without generating any original content.
No financial support was received for this study.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.