Abstract
The Jarisch–Herxheimer reaction (JHR) is a syndrome observed after antimicrobial treatment of some infectious diseases. The syndrome has clinical characteristics of an inflammatory reaction to antibiotic treatment. A prospective study of patients with a clinical and laboratory diagnosis of syphilis was conducted at a sexually transmitted diseases clinic in Rio de Janeiro, Brazil. Patients were treated with benzathine penicillin and observed for the JHR. A total of 115 patients were included in this study. Fifty-one patients (44%) had secondary syphilis; 37 (32%), primary; 26 (23%), latent; and one (1%), tertiary syphilis. Ten patients (9%) developed the JHR. All JHRs occurred in patients with secondary and latent syphilis. No patients experienced an allergic reaction to penicillin. The JHR occurred less frequently than in previous studies. It is important that health-care professionals recognize the clinical characteristics of the JHR so that it is not misinterpreted as an allergic reaction to penicillin.
INTRODUCTION
Sexually transmitted infections (STIs) are the fifth most common reason for medical consultations in Brazil. 1 Among all bacterial STIs, syphilis is the most prevalent in Brazil. In recent years, AIDS has been at the centre of discussion, but there has been a recent resurgence of syphilis among youth and links between HIV and syphilis are being drawn. In Brazil, syphilis ranks in fourth place among reportable diseases and remains a significant public health problem. According to the World Health Organization, there are approximately 900,000 new cases of syphilis in Brazil every year. 1
The Jarisch–Herxheimer reaction (JHR) is a syndrome observed after antimicrobial treatment of various infectious diseases, including syphilis. 2 It is characterized by fever, chills, malaise, headache, myalgias, a macular eruption and arthritis. This reaction can be confused with an allergic reaction to penicillin, which can lead to an erroneous diagnosis of penicillin allergy and to inappropriate termination of treatment, or patient's refusal of treatment based on past reactions.
As a curious clinical phenomenon noted after the introduction of syphilis treatment, the JHR was described for the first time by Jarisch in 1895 in Austria and Herxheimer in 1902 in Germany; at that time syphilis was treated with mercury.
Despite the fact that the JHR was first described more than one hundred years ago, neither the incidence nor the pathogenesis of the reaction is clear. Rosen et al. 3 reported rates of the JHR in 95% of patients with seropositive primary syphilis. He also reported the JHR rate to be 55–63% in those with seronegative primary syphilis, and 75% in patients with secondary syphilis. Griffin 4 quotes an incidence of the reaction in 80% of patients treated for early syphilis. Loveday 5 reported the rate to be 69% in a small test group of secondary syphilis patients. Overall, reports cite a high incidence of JHR after penicillin treatment of spirochete infections: 75–80% of those with syphilis, 54% of tick-borne relapsing fever patients and 82% of louse-borne relapsing fever patients. The JHR remains an important clinical issue in the treatment of syphilis and the exact incidence is a point of controversy. This study sets out to prospectively measure the incidence and remission of the reaction, by following the clinical evolution of syphilis patients after treatment.
METHODOLOGY
Enrolment
A prospective, descriptive study of 115 patients with clinical and laboratory diagnosed syphilis was conducted. All patients were enrolled in the Professor Rubem David Azulay Dermatology Institute, Santa Casa da Misericórdia in Rio de Janeiro, Brazil. The Institute's Ethics Committee approved the research. The study was carried out between March 2003 and March 2006.
Study population
The patients studied were of both sexes, at least 16 years old, and in any stage of the disease. Some were patients of the Institute's General Dermatology division, and others were referred from other private and public health providers in the city and the state of Rio de Janeiro. A number of patients were referred from the blood bank after positive serological test results. During the initial visit, the patients were examined following standard clinical procedures with special attention to mucocutaneous and genital areas. The clinical criteria used to classify the stages of the disease were taken from the Sexually Transmitted Diseases Treatment Guidelines 2002. 6 Briefly, patients with ulcers or chancres were given a diagnosis of primary syphilis. Patients with skin rash, mucocutaneous lesions and lymphadenopathy symptoms were classified as secondary syphilis. Patients with cardiac, ophthalmic, auditory abnormalities and gummatous lesions were diagnosed with tertiary syphilis. Asymptomatic seropositive patients were given a diagnosis of latent syphilis.
Inclusion criteria
In some patients the diagnosis was made on the basis of the clinical picture and was confirmed by serological tests. The Venereal Disease Research Laboratory (VDRL) test was considered reactive with a titre of 1:8 or greater and was confirmed by a positive fluorescent treponemal antibody absorption (FTA-Abs) test. While some patients did not present with a clinical picture of syphilis, they had tested positive for both VDRL and FTA-Abs tests at the blood bank and were diagnosed with latent syphilis. Patients agreeing to the study protocol signed an informed consent form.
Exclusion criteria
Those who reported a history of allergy to penicillin and those who did not agree to participate were excluded. Patients with a positive VDRL test for syphilis that was associated with previously treated infection were classified as having a ‘serological scar’ and were also excluded from the study.
Laboratory testing
The serum samples collected from the patients during screening were processed in the Santa Casa laboratory using the VDRL and FTA-Abs tests, both of them performed in accordance with manufacturer's instructions (bioMérieux, Marcy I' Etoile, France). An additional test, the Treponema pallidum haemagglutination assay (bioMérieux), was processed in the Bacteriology Department of Oswaldo Cruz Institute (FIOCRUZ) according to the manufacturer's directions. All patients were offered testing for HIV.
Protocol and clinical parameters
Patients diagnosed with syphilis received treatment in the Dermatology Sector in the Santa Casa da Misericórdia hospital. All collected data were recorded on standardized forms. The patient's stage of syphilis was assessed based on physical examination and on serology results using the criteria described previously. Following each penicillin injection, patients remained under surveillance in the ambulatory clinic for at least six hours. They were given information about possible side-effects, such as fever, worsening of lesions, headache and malaise. After the observation interval they were evaluated using a questionnaire, which included specific questions about the occurrence of signs and symptoms of the JHR, and how long after the injection they occurred. The patients were then instructed to return to the clinic in 72 hours for a re-evaluation. At the patient's follow-up visit, the attending physician further completed the study questionnaire by asking about additional manifestations of the reaction. The questionnaire investigated the possibility of worsened dermatological lesions (ulcerated lesions, rashes, mucous plaques, vesicular eruptions, syphilitic roseola, hair loss, etc.), malaise, myalgia, headache, prostration, fever, chills, tachycardia, tachypnea, pharyngitis, arthralgia, pruritus, burning, mental confusion and lymphadenopathy. Information on sexual practices, weight, temperature, blood pressure, blood tests and serological exam results was also recorded on the questionnaire.
Treatment
Patients were treated according to the 1999 Brazilian National Programme for STIs and AIDS (PN-DST/AIDS) from the Ministry of Health. Those with signs of primary syphilis were treated with one dose of Benzathine Penicillin G, 2.4 million units intramuscular. Patients diagnosed with secondary syphilis were treated with two doses of Benzathine Penicillin, 2.4 million units intramuscular each, separated by one week. Those with latent syphilis received three doses of Benzathine Penicillin G at one-week intervals, 7.2 million units total. 7
RESULTS
A total of 115 patients were selected to participate in the study: 81 were men (70%) and 34 were women (30%). The largest number of patients (51; 44%) presented with secondary syphilis. Of the remaining patients, 37 (32%) had primary syphilis, 26 (23%) latent syphilis and one (1%) had tertiary syphilis (Table 1). All 26 patients with latent syphilis were referred by the blood bank. Out of 115 study patients, 75 consented to HIV testing and six tested HIV-positive.
Characteristics of the 115 patients studied
The JHR was observed in 10 patients (9%). Men and women were affected at equal rates. It occurred in eight patients with secondary syphilis (n = 51), and two with latent syphilis (n = 26). It was not observed in patients with primary or tertiary disease. Two patients with HIV developed the JHR. In all patients, the reaction occurred after the first dose of penicillin.
The clinical symptoms exhibited by affected patients are detailed in Table 2. Fever was the most common manifestation. One (10%) patient developed a worsening of the rash of secondary syphilis. Symptoms typically started within three to six hours after treatment. All patients recovered spontaneously, thus no additional treatment was required. None of the patients in this study experienced an allergic reaction to penicillin.
Clinical manifestations of JHR in 10 patients affected
JHR = Jarisch–Herxheimer Reaction
DISCUSSION
Syphilis is a complex disease, generally contracted through sexual contact with infected lesions or bodily fluids. Transmission across the placenta from mother to child occurs and results in congenital syphilis. Transmission through a blood transfusion or an inoculation accident is rare. The aetiological agent is a spirochete, pathogenic only for human beings.
The JHR involves an exacerbation of cutaneous lesions of the underlying disease and/or the occurrence of other systemic signs and symptoms following potent antimicrobial therapy. The reaction generally begins six to eight hours after treatment. The maximum temperature occurs after approximately seven hours and decreases 12–24 hours after the injection of benzathine penicillin. It was reported to be a very common syndrome observed after antimicrobial treatment of several infectious diseases caused by spirochetes. These include Lyme disease, leptospirosis, louse-borne relapsing fever, brucellosis and syphilis. 8 While rarely associated with long-term morbidity, the reaction can be symptomatically severe and consists of fever (average rise in temperature of 0.8–1.5°C), chills, malaise, headache, myalgias, tachycardia, tachypnea, pharyngitis, leukocytosis (total white blood cell count of 12,500/mm3), hypotension, vasodilatation, lymphadenopathy, arthritis and an urticarial macular eruption. A vesicular eruption has also been described, peculiarly in four African-American patients, 3 however, we have not observed this phenomenon. During the fever, other signs of the disease can become more evident, for example, the faint macular rash of secondary syphilis may become more apparent. In cardiovascular syphilis there is a risk of ostial oedema with subsequent coronary occlusion or rupture of an aneurysm. 8
The pathogenesis of the JHR is still not clear. It was thought to be caused by the release of endotoxins as treponemes are killed and therefore the reaction would be severe proportionally to the number of treponemes, but others claim that the number of bacteria has no effect on the probability or severity of the reaction. 2,9 In one study, endotoxins were not detected in any of the patients who developed the JHR. 4,10 Endotoxins present among some of the JHR patients were probably a result of sepsis related to endogenous Gram-negative bacteria. These observations tend to refute the theory that endotoxin plays a role in the reaction. A more likely explanation involves the cytokine cascade. Studies have shown a rise and fall of tumour necrosis factor (TNF), interleukin-6 (IL-6) and interleukin-8 (IL-8) after the penicillin injection. 4,11–13 A recent study of relapsing fever patients exhibiting symptoms of the JHR after penicillin treatment showed a massive release of cytokines in the bloodstream. 12 TNF was identified even before clinical symptoms occurred. IL-6 appeared at the beginning of the symptoms and IL-8 with the onset of fever.
The frequency of the JHR might have been overestimated in the medical literature. According to previous studies around 80% of syphilis patients experienced the JHR while we witnessed the reaction in only 9% of those treated. In stark contrast with previously reported JHR incidence rates of 55–95% in primary syphilis patients, we observed no JHR in patients with primary syphilis. Clinicians who have little experience with JHR may not recognize this syndrome and may confuse the JHR response with an allergic reaction to penicillin. This can be a particularly serious problem due to the widespread notion among patients in Brazil that any allergic reaction to penicillin can be fatal. Patients claim to be allergic to penicillin to avoid treatment. At present, penicillin is the most effective medicine to treat syphilis. If the JHR is recognized and properly diagnosed, misconceptions about penicillin allergy can be avoided and penicillin can be used.
This study is limited by difficulties in measuring follow-up observations. All patients were observed for six hours after penicillin treatment, and though maximum temperature of the reaction typically occurs within six to eight hours, it is possible that all reactions were not observed. After 72 hours, patients were reassessed by questionnaire and re-examination. Furthermore, blood pressure and body temperature data were incomplete, preventing a more detailed analysis of the reaction. If more data were collected or if the patients were monitored for a longer period, the incidence of the JHR might have been found to be higher.
Even though the incidence of the JHR is low, it is important that patients and health-care workers understand the difference between the JHR and allergic reaction to the antibiotics. Clinicians should educate patients to expect the JHR and should explain the benign nature of the symptoms in order to avoid future non-compliance with syphilis treatment.
Footnotes
Acknowledgements
We thank Hsi Liu PhD from the Centers for Disease Control and Prevention and Thomas Miller MD from the University of North Carolina at Chapel Hill School of Medicine for their assistance in the revision of the paper.