Abstract
Nervous system involvement is a rare manifestation of brucellosis. We describe our experience of the diagnosis, treatment and final outcome of patients with neurobrucellosis at the Erciyes University Gevher Nesibe Hospital, a tertiary referral centre in Central Anatolia, Turkey. Thirty-six adult patients were diagnosed with neurobrucellosis from January 1997 to December 2006. Headache and fever were the most common symptoms. Neck stiffness was present in 25 patients. Brucella spp was isolated from the blood of nine patients and from the cerebrospinal fluid of 11. Doxycycline (by mouth) plus rifampin (by mouth) with ceftriaxone (intravenously) were the most common treatment choices. Three patients died as a result of problems other than neurobrucellosis and relapse occurred in one patient. Neurobrucellosis presents with hetoregenous clinical signs.
Introduction
Involvement of the nervous system has been detected in approximately 2–6.5% of all brucellosis cases. 1 We describe our experience of the diagnosis, treatment and final outcome of patients with neurobrucellosis who were diagnosed and treated at the Erciyes University Gevher Nesibe Hospital, a 1300-bed tertiary referral centre in Central Anatolia, Kayseri, Turkey.
Patients and methods
We retrospectively reviewed the records of the Department of Infectious Diseases in order to identifiy patients aged ≥16 years with neurobrucellosis treated at the hospital from January 1997 to December 2006. The demographic and clinical information were extracted from the patients' charts. The cases were classified as acute (less than eight weeks), sub-acute (from eight to 52 weeks), or chronic (more than one year), according to the duration of symptoms. 1 The persistence of signs or symptoms of the disease at the end of therapy was considered as a therapeutic failure. Relapse was defined as an occurrence of similar symptoms and signs after the completion of therapy and/or another positive culture. 2
Neurobrucellosis was diagnosed using the following criteria: symptoms or clinical findings consistent with central nervous system infection; isolation of Brucella spp from the cerebrospinal fluid (CSF); and/or finding antibodies to Brucella spp in the CSF (at any titre); the presence of any abnormality in the CSF (pleocytosis, increased protein levels or decreased glucose levels) with a positive serology of standard tube agglutination test (STA) ≥1:160) for brucellosis.
Results
During a ten-year period 917 patients with brucellosis were admitted to the department. Thirty-six (3.9%) were diagnosed with neurobrucellosis. Clinical forms of the disease were acute in 26 cases, sub-acute in eight and chronic in two. In 14, the duration of the complaints were of less than 14 days. Meningitis was the most common manifestation.
Headache and fever were the most common symptoms and neck stiffness was present in 25. At least, one of the findings, e.g. headache, fever or neck stiffness, were present in over 90%. The demographic and clinical features of the patients are given in Table 1.
Demographic and clinical features of the patients with neurobrucellosis
*Doxycycline (100 mg by mouth every12 h) plus rifampin (600-900 mg/day by mouth)
†Ceftriaxone (2 g intravenously every12 h) was initially added to the regimen for 2–3 weeks
‡Streptomycin (1 g/day intramuscularly)
Brucella spp was isolated from the blood in six patients, from the CSF in eight and from both CSF and blood in three. Abnormal computerized tomography scan findings, including paranchimal oedema (six) and cerebral infarct (one), were detected in seven patients. The laboratory results are shown in Table 2.
Laboratory findings of patients with neurobrucellosis
ESR, erythrocyte sedimentation rate; WBC, white blood cell count; CRP, C-reactive protein; CSF, cerebrospinal fluid; ALT, alanine aminotransferase; STA, standard tube agglutination test; CT, computerized tomography
Nervus abducens involvement was detected in two patients. One recovered but the other persisted despite treatment. Motor deficiencies were detected in the lower extremities in five patients and in the upper extremities in two. One patient received physical therapy due to persistence of the symptoms. Hemiparesis was detected in two patients and improved in one. Urinary and gaita incontinence was observed in two patients - one recovered.
Twenty-six patients were treated with a combination of doxycycline plus rifampin with ceftriaxone. A combination of doxycycline plus rifampin with streptomycin was initially given to two patients but streptomycin was discontinued due to the occurance of tinnitus after the tenth day of treatment. A clinical response and the return of CSF findings to normal values for the duration of the treatment were our main criteria. The treatment continued for three months in 33 patients and four months in three. For one patient doxycycline was changed to trimethoprim-sulfamethoxazole (160/800 mg by mouth every 12 hours) after two months because of photosensitivity. Rifampin-related hepatic toxicity was observed in one patient and, therefore, changed to trimethoprim-sulfamethoxazole.
The overall mortality was 8.3%. Three patients died from problems other than neurobrucellosis. Patients were followed up for 12 months after completion of the therapy. Relapse occurred in one patient. This patient had a ventriculoperitoneal shunt for pseudotumour cerebri and was treated for three months with doxycycline and rifampin; she completely recovered clinically. She was admitted to our hospital after four months of headache and diplopy after the end of therapy. Brucella spp. grew in her blood and CSF cultures. Doxycycline and rifampin with ceftriaxone was initiated and the ventriculoperitoneal shunt was extracted. During the follow-up her intracranial pressure did not increase. Ceftriaxone was stopped after three weeks. The doxycycline and rifampin treatment was continued for four months when both clinical and microbiological cures were achieved.
Discussion
In this retrospective study the rate of neurobrucellosis was 3.9% and the majority of patients presented with acute or sub-acute forms of the disease. However, a higher rate of neurobrucellosis (17.8%) has been previously reported from Turkey where chronic meningitis was more common. 3 This could have been a result of the different sample size.
Yetkin et al., 4 reported fever and headache as the major presenting complaints and observed in 85% and 70% of the patients, respectively. Heper et al., 5 reported similiar rates of headache and neck stiffness. Our results were concordant with those studies and suggest that CSF should be examined in all patients with brucellosis who present with any neurological symptoms including unexplained headaches.
Other authors have reported CSF positive culture rates of 15% to 30%. 3–5 The duration of the disease may play an important role in the isolation of Brucella spp. from the clinical specimens. In our study, the duration of symptoms was less than 14 days in 37.8% of the patients.
One of the problem in diagnosing neurobrucellosis is the difference of the type and duration of the antimicrobial treatment. In a previous study from our clinic, a combination of ceftriaxone with doxycycline and rifampicin for the treatment of neurobrucellosis had a beneficial effect. 6 Trimethoprim-sulfamethoxazole has been used in other studies. 7,8
In our study the mean duration of treatment was three months. The duration of the therapy should be decided on a case by case basis. Longer courses of therapy may be necessary for patients whose clinical and neurological improvment is slow and who have suffered the symptoms over a longer period. 8–10
Physicians should be aware that neurobrucellosis presents with heterogenous clinical signs in endemic areas.
Footnotes
Acknowledgements
This study was presented at the 18th European Congress of Clinical Microbiology and Infectious Diseases, Poster Session V, 21 April 2008, Barcelona, Spain (poster No. 1865, entitled Brucellosis).