The incidence of syphilis has increased, primarily among men who have sex with men. Syphilis presentation can be difficult to discern and might be missed, but should be suspected in persons who experience the classic Jarisch-Herxheimer reaction after treatment with β-lactam antibiotics. This case describes a 55-year-old male patient with negative test results for HIV who presented to the clinic as a contact of a recent male sexual partner who was diagnosed with gonorrhea. This case suggests that ceftriaxone and azithromycin treatment of persons with unknown infectious syphilis infections might cause a classic posttreatment syphilis reaction (rigour that spontaneously resolves). Patients should be informed to look for these Jarisch-Herxheimer reaction symptoms, and clinicians should assess for the symptoms of this reaction in patients who are at risk of syphilis.
Case
Visit 1. A 55-year-old man presented to the sexually transmitted infection (STI) clinic after a recent casual male partner informed him he had been diagnosed with gonorrhea. The patient engaged in condomless receptive and penetrative oral and anal sex with this man. Table 1 provides a summary of the patient’s case.
Review of systems: The patient was asymptomatic when he presented for care, but reported having had a painless genital lesion that was less than 1 cm2 approximately 4 to 6 weeks earlier; it resolved spontaneously after 7 to 10 days. The patient suspected the lesion was an “ingrown hair.” He denied rashes, hair loss, and mucous lesions. He denied current or recent rigour, fatigue, weight loss, lymphadenopathy, myalgia, arthralgia, headaches, and vision or hearing changes.
Past medical history: The patient’s past medical history included insomnia, gastroesophageal reflux disease, and hypertension, for which he took trazodone, esomeprazole, and the telmisartan-amlodipine combination, respectively. He had been taking these drugs for longer than 12 months. He had started taking a fixed-dose combination of emtricitabine (200 mg) and tenofovir (300 mg) daily for HIV preexposure prophylaxis (PrEP) 5 months earlier.
The patient had similarly presented to the STI clinic 5 months previously as a contact of a gonorrhea case and was treated with 1 intramuscular dose of 250 mg of ceftriaxone (reconstituted with 0.9 mL of 1% lidocaine) plus 1 oral dose of 1 g of azithromycin.1,2 This treatment was administered at the clinic. He experienced no reaction after treatment (both immediately after treatment and during the following day): no nausea, no emesis, no diarrhea, and no rigour. His test results at that time were negative for gonorrhea and chlamydia (urine nucleic acid amplification testing [NAAT], and pharyngeal and rectal cultures). He did not undergo serology testing at the clinic, but had documented negative HIV and syphilis test results from this time.
On examination: The patient was afebrile and had no perceptible rashes on his hands, feet, or trunk. His cervical nodes were not palpable or tender; he had no oral lesions or erythema. He had no palpable inguinal nodes or tenderness, and no lesions, erythema, or tenderness of his penis or genital area; no urethral discharge was present. He had no scrotal lesions or testicular tenderness. He had no external anal erythema, lesions, or discharge. Anoscopy was not performed.
Plan: The nurse practitioner who saw this patient at his initial visit collected pharyngeal and rectal swabs and urine for gonorrhea and chlamydia testing, as well as blood for HIV and syphilis testing. As the patient was a contact of a gonorrhea case, this same provider treated the patient in the clinic with 1 intramuscular dose of 250 mg of ceftriaxone (reconstituted with 0.9 mL of 1% lidocaine) plus 1 oral dose of 1 g of azithromycin.1,2 The patient remained in the clinic after the injection without reaction.
A fourth-generation antigen-antibody combination assay was used for HIV testing.3 Syphilis testing was done using the reverse screening algorithm, starting with a chemiluminescent microparticle immunoassay (CMIA), followed by a rapid plasma reagin (RPR) test and a Treponema pallidum passive particle agglutination assay (TPPA) for samples with positive screening results.4,5 Gonorrhea and chlamydia samples underwent NAAT.6–8 Of note, between the patient’s previous and current presentation for care, in Ontario, NAAT for extragenital samples was validated by the Public Health Ontario laboratory. This occurred owing to a greater than 2-fold increase in sensitivity of NAAT compared with culture, and because the test swabs required were reduced from 2 swabs to 1. Moreover, extragenital testing is done with the same type of swab used for endocervical gonorrhea and chlamydia testing and is thus likely more readily available in many clinics.
Test results: The patient had negative results for HIV, pharyngeal and urine gonorrhea and chlamydia, and rectal chlamydia. His rectal gonorrhea test result was positive. For syphilis, the CMIA result was reactive, the RPR titre was 1:4, and the TPPA result was reactive. The STI clinic nurses contacted the patient and requested that he return to the clinic.
Visit 2. On returning to the clinic 8 days after the first visit, the patient was asymptomatic. To determine the need for gonorrhea re-treatment, I inquired if he had had issues with treatment. He denied nausea, emesis, and diarrhea, and he denied sexual contact with untreated partners. He reported rigour about 2 hours after receiving gonorrhea treatment; rigour lasted less than 12 hours and resolved spontaneously. He denied rashes, mucosal irritation and pain, myalgia, and arthralgia. He denied experiencing such symptoms when he was treated empirically for gonorrhea 5 months earlier.
I repeated the syphilis bloodwork, determined the patient’s stage of infectious syphilis (early latent phase owing to no symptoms, a confirmed negative result 5 months earlier, and no known contacts with any sexual partners recently diagnosed with infectious syphilis). I treated him with 1 intramuscular dose of 2.4 million units of benzathine penicillin G.1,2 He tolerated the injection well and had no reactions during the 15 minutes he remained in the clinic. Interestingly, he reported a similar Jarisch-Herxheimer–like reaction after this treatment as well. Patients who receive more than 1 dose of antibiotics for syphilis typically only experience the Jarisch-Herxheimer reaction with the first treatment; it is possible that the second reaction occurred in this case because, in the first instance, the nurse practitioner had not actually treated him for syphilis. Instead, the nurse practitioner had treated him for gonorrhea, and potentially induced this reaction with a dose of medication that was appropriate for gonorrhea but subtherapeutic for syphilis. It is thus possible that he experienced the Jarisch-Herxheimer reaction twice.
Syphilis bloodwork from this second visit revealed reactive CMIA results, an RPR titre of 1:8, and reactive TPPA results, supporting the diagnosis of infectious syphilis. The change in syphilis titre might suggest the infection was primary (with a possible chancre in an undetected location such as the rectum); however, it might also be normal laboratory variation in the measurement of a serofast state.
Discussion
It is possible this patient experienced a Jarisch-Herxheimer reaction, which is common after treatment of spirochete infections (eg, syphilis, yaws, pinta, Lyme disease).1,2,9 This reaction typically starts 2 hours after treatment and resolves within 24 hours; it can occur during any stage of syphilis.1,2,9 It is hypothesized that this reaction is due to mass release of lipoproteins from destroyed bacteria that induce rigour, myalgia, arthralgia, and headache.9 The Jarisch-Herxheimer reaction is not a hypersensitivity, and the patient does not need to avoid penicillin. No treatment is required, although antipyretics can alleviate symptoms.2 Systemic corticosteroids can be used for severe reactions, but only with expert consultation.2
The postulation here is that this patient experienced such a reaction after receiving treatment for gonorrhea with a cephalosporin and macrolide. This is fitting, seeing as cephalosporins have a similar mechanism of action to penicillin,10 and because these 2 medications can cure syphilis, although in higher doses and for a longer duration than what this patient received.1,2 It is possible that such a reaction occurred here from a subtherapeutic dose of these drugs. Supporting this assertion is that the patient had a similar reaction when treated with benzathine penicillin G.
One similar case report exists. In this other case, the patient was an HIV-positive 32-year-old man who, similar to this case, was diagnosed with rectal gonorrhea, was treated with ceftriaxone and azithromycin, and experienced rigour 6 hours later.11 The patient returned to the clinic with a classic syphilis rash and received treatment for syphilis. Serology results supported the syphilis diagnosis.
Another possible explanation for this patient’s symptoms is HIV seroconversion,2 although this is unlikely based on his use of HIV PrEP, which can prevent HIV transmission in more than 90% of those who take it.12 The patient’s symptoms might also have been the result of an unrelated concomitant viral infection (eg, influenza), although the timing and duration of symptoms makes this unlikely. Likewise, drug-drug interactions are an unlikely cause of the patient’s symptoms, as the only possible interaction is a category C interaction between azithromycin and trazodone (potential corrected QT interval prolongation),13 and he had no flulike reaction with concurrent use of these medications 5 months previously. The patient’s reaction might also have been a Jarisch-Herxheimer reaction to a different spirochete, such as Borrelia burgdorferi, which is prevalent locally.14 No testing was done to rule out Lyme disease, so this infection is possible, although the patient did not have any related symptoms including erythema migrans or neurologic findings.15 As I work in an STI clinic, testing for Lyme disease is not available. Instead, I encouraged the patient to follow up with his family physician for further assessment.
Recommendations for practice
This case highlights 3 points. The first is the need for clinicians to include syphilis in the differential diagnosis of oral, genital, and perianal lesions.1,2 This is particularly important owing to increasing rates of syphilis, primarily among men who have sex with men.16,17 In such cases, it is ideal to consider (and provide) empiric treatment at the point of care, plus appropriate testing including serology and the consideration of direct fluorescent antibody (DFA) or polymerase chain reaction (PCR) testing of syphilitic lesions.1,2 Specifically, DFA and PCR testing involve specimen collection from a potential syphilitic lesion, whether a chancre, condyloma latum, or mucous patch. When results are positive, DFA and PCR confirm the presence of syphilis organisms. Of note, DFA and PCR testing of syphilis lesions can detect primary infection before the development of systemic markers that can be detected in serology.
Second, the symptoms of Jarisch-Herxheimer reaction should be communicated to patients who are at risk of syphilis who receive treatment for gonorrhea with ceftriaxone and azithromycin. This involves explicitly listing these symptoms to patients at the time of treatment as part of reviewing posttreatment precautions (eg, reviewing the symptoms of anaphylaxis, recommending avoiding sexual activity until no longer infectious, recommending avoiding sexual contact with untreated partners). Patients should be instructed to return to the clinic for assessment if they experience Jarisch-Herxheimer–like symptoms, and clinicians should consider providing empiric treatment while investigations are pending for patients with these symptoms. This approach aligns with the previous case report,11 in which the clinicians aptly suspected and empirically treated syphilis based on the patient’s risk factors for syphilis plus a Jarisch-Herxheimer–like reaction after receiving ceftriaxone and azithromycin treatment for gonorrhea. Similarly, clinicians who examine patients who were recently treated with these medications should also explicitly inquire about Jarisch-Herxheimer reaction symptoms, and not assume that patients would necessarily report such symptoms without explicit inquiry. Although the patient in this case volunteered this information without being precisely asked about the symptoms of this reaction, he only provided this information once I inquired if he had experienced any symptoms after his gonorrhea treatment. This highlights the need for clinicians to positively review these symptoms.
Third, although less applicable to this case because the patient was already taking HIV PrEP, syphilis is an established risk factor for HIV acquisition, meaning that a syphilis diagnosis should signal clinicians to ensure HIV testing is performed and, if test results are negative for HIV, to consider HIV PrEP. In the existing PrEP studies,18 seroconversion rates within 12 months of syphilis diagnosis ranged between 1 in 20 and 1 in 30 persons.19,20 Data from Vancouver, BC, also found elevated HIV incidence after syphilis diagnosis (3.6 per 100 person-years), which increased to 17 per 100 person-years for patients with concurrent gonorrhea and syphilis diagnoses.21 Thus, nearly 1 in 5 such persons would acquire HIV within 12 months of this presentation, highlighting the importance of PrEP for such patients. Recent Canadian guidelines12 detail how to provide this intervention.
Conclusion
This article reviews the case of an asymptomatic 55-year-old man with negative test results for HIV who presented as a contact of a gonorrhea case, experienced rigour after ceftriaxone and azithromycin administration, and was subsequently diagnosed with syphilis. This case supports a previous case report of a similar situation,11 and highlights that clinicians should inform patients about Jarisch-Herxheimer reaction symptoms and consider these symptoms as indicators of syphilis in otherwise asymptomatic patients. Finally, clinicians should discuss HIV PrEP with patients diagnosed with syphilis, considering the elevated HIV seroconversion rates that occur after this diagnosis. This helps ensure comprehensive sexual health service provision.
Notes
Editor’s key points
▸ The incidence of syphilis has increased, primarily among men who have sex with men. This article reviews the case of an asymptomatic 55-year-old man with negative test results for HIV who presented as a contact of a sexual partner with gonorrhea, experienced rigour after ceftriaxone and azithromycin administration, and was subsequently diagnosed with syphilis.
▸ Syphilis should be suspected in persons who experience the classic Jarisch-Herxheimer reaction (rigour that spontaneously resolves) after treatment with β-lactam antibiotics. The symptoms of Jarisch-Herxheimer reaction should be communicated to patients who are at risk of syphilis who receive treatment for gonorrhea with ceftriaxone and azithromycin. Patients should be instructed to return to the clinic for assessment if they experience these symptoms, and clinicians should consider providing empiric treatment while investigation results are pending.
▸ Clinicians should discuss HIV preexposure prophylaxis with patients diagnosed with syphilis, considering the elevated HIV seroconversion rates that frequently occur after this diagnosis.
Points de repère du rédacteur
▸ L’incidence de la syphilis a augmenté, surtout chez les hommes qui ont des relations sexuelles avec des hommes. Cet article passe en revue le cas d’un homme asymptomatique de 55 ans dont les résultats de dépistage du VIH étaient négatifs. Il avait consulté en raison d’un contact sexuel avec une personne atteinte de gonorrhée, et il a eu un frisson solennel à la suite de l’administration de ceftriaxone et d’azithromycine. Il a par la suite reçu un diagnostic de syphilis.
▸ Il y a lieu de suspecter la syphilis chez les personnes qui ont la réaction classique de Jarisch-Herxheimer (frisson solennel qui disparaît spontanément) après une thérapie aux antibiotiques β-lactame. Il faut expliquer les symptômes de la réaction de Jarisch-Herxheimer aux patients à risque de syphilis qui reçoivent un traitement pour la gonorrhée avec de la ceftriaxone et de l’azithromycine. Il faut aviser les patients qu’il faut revenir à la clinique pour une évaluation s’ils présentent de tels symptômes, et les cliniciens devraient envisager un traitement empirique en attendant les résultats de l’investigation.
▸ Les cliniciens devraient discuter de la prophylaxie préexposition au VIH avec les patients ayant reçu un diagnostic de syphilis, étant donné les taux élevés de séroconversion contre le VIH qui se produisent fréquemment après un tel diagnostic.
Footnotes
Competing interests
None declared
This article has been peer reviewed.
Cet article a fait l’objet d’une révision par des pairs.
- Copyright© the College of Family Physicians of Canada