Efficacy of prolonged antimicrobial chemotherapy for brucellar spondylodiscitis

Transcription

1 ORIGINAL ARTICLE INFECTIOUS DISEASES Efficacy of prolonged antimicrobial chemotherapy for brucellar spondylodiscitis S. Ioannou 1, D. Karadima 2, S. Pneumaticos 3, H. Athanasiou 4, J. Pontikis 5, A. Zormpala 6 and N. V. Sipsas 1 1) Pathophysiology Department, Laikon General Hospital, and Medical School, National and Kapodistrian University of Athens, Athens, 2) Internal Medicine, Panarkadikon Hospital, Tripolis, 3) Orthopedics Department, KAT Hospital, and Medical School, National and Kapodistrian University of Athens, Athens, 4) Internal Medicine Department, Arta General Hospital, Arta, 5) Radiology Department, Panarkadikon Hospital, Tripolis and 6) Radiology Department, Laikon General Hospital, and Medical School, National and Kapodistrian University of Athens, Athens, Greece Abstract The standard treatment of brucellar spondylitis with a combination of two antibiotics for 6 12 weeks is associated with high rates of treatment failure and relapse. The present study aimed to assess the safety and efficacy of a treatment strategy based on the prolonged administration of a triple combination of suitable antibiotics. Eighteen patients with brucellar spondylitis were treated with a combination of at least three suitable antibiotics (doxycycline, rifampin, plus intramuscular streptomycin or cotrimoxazole or ciprofloxacin) until the completion of at least 6 months of treatment, when clinical, radiological and serology re-evaluation was performed. If necessary, the treatment was continued with additional 6-month cycles, until resolution or significant improvement of clinical and radiological findings, or for a maximum of 18 months. At presentation, the median age was 66 years (range, years) with male predominance. The median duration of therapy was 48 weeks (range weeks). Treatment was discontinued early because of side-effects in only one patient. Surgical intervention was required for three patients. At the end of treatment all patients had a complete response. After completion of treatment, all patients were followed up with regular visits. During the follow-up period (duration 1 96 months, median 36.5 months), no relapses were observed. In conclusion, prolonged (at least 6 months) administration of a triple combination of suitable antibiotics appears to be an effective treatment for brucellar spondylitis. Keywords: Brucella, spondylodiscitis, treatment, vertebral Original Submission: 10 February 2010; Revised Submission: 26 April 2010; Accepted: 13 May 2010 Editor: G. Pappas Article published online: 27 May 2010 Clin Microbiol Infect 2011; 17: /j x Corresponding author: N. V. Sipsas, Pathophysiology Department, Medical School, University of Athens, Mikras Asias 75, Athens, , Greece nsipsas@med.uoa.gr Introduction Brucellar spondylitis might be a devastating complication because, quite often, it is associated with neurological complications, requires spinal surgery, and results in permanent functional sequelae [1]. Even though it was described back in 1932 [2], there is still controversy regarding major treatment issues, such as the optimal antibiotic regimen, the duration of treatment and the criteria by which to consider the disease as cured [3]. The reason for this confusion is that the existing studies comprise small series of cases, treated and followed very heterogeneously. In the largest published studies [4 8], various combinations of two antibiotics with known efficacy against Brucella were administered for 6 12 weeks, with disappointing results. Alp and coworkers reported that the standard regimen (doxycycline combined with streptomycin) for at least 12 weeks remains the first choice [3]. Solera et al treated 35 patients for a median duration of antimicrobial therapy of 120 days, but with high relapse and failure rates (5/35, and 9/ 35 respectively) [9]. Nevertheless, all these series and a recent meta-analysis [10] made clear that failure rates are much higher when the duration of treatment is 6 weeks or less than when treatment is continued until 12 weeks (43.66% vs %). Therefore the duration of treatment suggested by the World Health Organization for acute, uncomplicated brucellosis (6 weeks) is clearly not sufficient for patients with spondylitis. Yet, even 12 weeks of treatment Clinical Microbiology and Infection ª2010 European Society of Clinical Microbiology and Infectious Diseases

2 CMI Ioannou et al. Brucellar spondylodiscitis treatment 757 is not optimal; a recent retrospective study of 96 patients treated for at least 3 months with suitable combinations of antibiotics reported treatment failure rate 20%, attributable mortality 2.6% and 6% of the patients had severe functional sequelae [7]. On the other hand, there are some small, early studies, suggesting that a triple combination of suitable antibiotics administered for 6 8 weeks achieves more acceptable cure rates [11,12]. The high failure and relapse rates observed in patients treated for 6 12 weeks urged three Greek centres to adopt a different treatment and follow-up strategy for brucellar spondylitis based on the several principles: the minimum duration of treatment should be 24 weeks; if there is no clear resolution of clinical and radiological findings the treatment should be continued for up to 72 weeks; the regimen should contain three suitable antibiotics; a complete clinical, serology and imaging work-up should be performed every 6 months, to rule out relapse and to assess the need for further treatment. In the present study, we report the efficacy of this treatment strategy in 18 consecutive patients with brucellar spondylitis. Materials and Methods Study population and setting From October 2000 until December 2009, 11 cases of brucellar spondylitis were diagnosed and followed up prospectively at the Laikon General Hospital, Athens; one case at the General Hospital of Arta, a city in north-western Greece; and six cases at the General Hospital of Tripolis, a city which lies in southern Greece; the last hospital serves an area with a long tradition of sheep farming. Diagnosis The diagnosis of brucellosis was established by the clinical findings compatible with brucellosis, positive standard agglutination test (SAT, 1:160 or higher) [13], and/or by isolating Brucella species from blood or bone marrow, other body fluids, or tissue specimens. An ELISA for Brucella-specific IgM and IgG serum antibodies, as well as the Coombs test, were used as additional diagnostic tools where available. In five cases, computed tomography-guided biopsy of the affected spinal region was performed. Demographics and laboratory evaluation The demographic variables recorded at presentation were age, sex, occupation, type of exposure to Brucella melitensis, associated illnesses, clinical presentation and diagnostic delay. The baseline haematology studies included total and differential white blood cell counts, haemoglobin levels, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) and biochemistry profile. Radiographic studies Diagnosis of spondylitis was confirmed by magnetic resonance imaging (MRI). Findings suggestive of spinal involvement were decreased signal intensity in the vertebral bodies on T1-weighted images, increased signal in the vertebral bodies on T2-weighted images, increase in signal in discs on T2-weighted images, loss of end-plate definition on T1-weighted images and contrast enhancement in the discs on T1-weighted images with gadolinium [14]. Definitions Spondylitis was defined as infection of the intervertebral disk and the adjacent vertebrae, with or without associated epidural or psoas abscesses [15]. Therapeutic failure was defined as persistent or worsening symptoms and signs of the disease and increasing MRI findings, at the completion of 24 weeks of therapy [4]. Partial remission was considered to have clinical improvement and stable or improved MRI findings (i.e. no new lesions and stable or improved initial lesions). Complete remission was considered as the resolution of all clinical findings (with the exception of mild residual mechanical ) and resolution or significant improvement of radiological findings of spondylitis. Relapse was defined as the reappearance of symptoms, new positive blood cultures, and/or worsening imaging findings during the follow-up, after the discontinuation of antibiotics [4]. Sequelae were considered to have occurred when, neurological deficits or functional limitation persisted for longer than 6 months after the end of treatment. The severity of clinical sequelae was categorized according to the patient s functional status: mild sequelae, no neurological deficits but with exercise, not preventing the patients from performing their usual work or daily activities; moderate sequelae, mild neurological deficits and, interfering with patient s ability to work; and severe sequelae, permanent requiring bed rest and analgesics, and/or motor or sensorial deficits [4]. Indications for surgery were severe mechanical instability of the affected spine, treatment failure after 24 weeks of antibiotic therapy, large paravertebral or epidural abscesses not responding to medical treatment, cauda equina syndrome, and severe weakness of muscle as a result of collapse of the vertebral body [4,9]. Adverse effects that could result in treatment modification included: hepatotoxicity, defined as a more than five-fold

3 758 Clinical Microbiology and Infection, Volume 17 Number 5, May 2011 CMI TABLE 1. Treatment and follow-up strategy Week Intervention 0 Diagnosis External immobilization of the spine 1 3 Doxycycline (per oral, 100 mg twice daily) plus rifampin (900 mg, per oral once daily) plus streptomycin (1 g i.m., once daily) 4 24 Doxycycline (per oral, 100 mg twice daily) plus rifampin (900 mg, per oral once daily) plus sulfamethoxazole trimethoprim (800/160 mg, p.o., twice daily) a Follow-up visits Clinical examination Routine haematology and biochemistry to rule out side-effects of the treatment 24 Re-evaluation Clinical Haematology biochemistry Blood culture Serology SAT ELISA MRI scan of the affected spine Complete remission Partial remission Treatment failure Discontinuation of treatment, follow-up every 6 months for up to 3 years Clinical SAT MRI (every year) Continuation of treatment until completion of 48 or 72 weeks as needed Follow-up every 6 months for up to 3 years Consideration of surgical therapy Debridement Diskektomy Drainage of abscesses Change of the antibiotic regimen, according to susceptibility tests, when available SAT, standard agglutination test; MRI, magnetic resonance imaging. a Ciprofloxacin (500 mg, per oral, twice daily) was reserved as a second line option in case of discontinuation of rifampin or sulfamethoxazole trimethoprim. increase in aspartate or alanine aminotransferase levels over the baseline value or more than ten times the upper limit of the normal range; nephrotoxicity, defined as more than twice the upper limit of the normal creatinine level range; ototoxicity; serious skin reactions; and central nervous system toxicity [16]. Treatment and follow-up All patients were treated and followed up using the protocol described on Table 1. In December 2009, all patients were contacted and re-evaluated by the caring physicians. The Ethics Committees of the participating hospitals approved the study protocol and all patients provided their informed consent. Results Baseline characteristics During the study period, 18 patients (12 were male) with brucellar spondylitis were diagnosed, treated and followed up according to our protocol. The median age was 66 years (range years) (Table 2). Four patients had an occupational exposure and all the patients reported ingestion of unpasteurized dairy products. The vertebral level involved was lumbar or lumbo-sacral in 11 (61.1%) cases, cervical in two (11.1%) cases, thoracic in one (5.5%) case, whereas, in four (22.2%) cases, lumbar and thoracic regions were affected simultaneously (Table 2). At presentation, all patients experienced, either constant or occasional, of the affected spinal region. The time from the onset of symptoms to the diagnosis of spondylitis was in the range days (median 45 days). For 14 patients (77.7%), the diagnoses of brucellosis and spinal osteomyelitis were made simultaneously. In four patients, vertebral involvement was diagnosed 2, 3, 7 and 9 months after the diagnosis of brucellosis, respectively. In these four patients, standard therapy with two antibiotics was given prior the diagnosis of spondylitis, for various time periods. At presentation, the most common laboratory finding was elevated ESR (100%) and elevated CRP levels (100%). ESR was in the range mm/h (median 46.5 mm/h). CRP levels were in the range mg/l (median, 8.5 mg/l). Anaemia was found in ten patients (55.5%) and leukocytosis in four patients. Abnormal liver function tests were found in seven patients (38.8%), which were attributed to brucellosis. Standard agglutination testing of initial samples was positive in 13 (72.2%) of the 18 patients. In the rest of the patients with negative SAT, computed tomography-guided biopsy of the

4 CMI Ioannou et al. Brucellar spondylodiscitis treatment 759 TABLE 2. Demographics, and baseline clinical, laboratory, and imaging findings among the 18 study patients with brucellar spondylitis Patient Sex Age (years) Vertebral level affected Epidural and/or paravertebral masses Findings at diagnosis Clinical SAT titres MRI 1 Male 67 L5 S1 ()) Constant low back, fever 1: Male 74 L4 L5 ()) Severe, fever 1: Male 48 C3 C4 ()) Neck, night sweats 1: Female 67 L2 L3 (+) Severe, night sweats <1: Male 85 L2 L3 (+) Lumbar, fever <1: Female 61 L2 L3 (+) Severe, constant <1: Male 42 C3 C4 (+) Neck <1: Male 67 L3 L4 (+) Lumbar, fever, night sweats 1: Male 65 L5 S1 ()) Low back, fever 1: Female 70 L4 L5 (+) Constant low back 1: Male 65 L2 L3 (+) Lumbar, fever 1: Male 45 T11 T12 ()) Lumbar, fever 1: L2 L3 13 Female 67 T9 T10 (+) Constant, fever 1: Female 64 L2 L3 ()) Lumbar <1: Male 76 T6 T7 ()) Constant, fever 1: L4 L5 16 Female 81 L3 L4 ()) Lumbar, fever 1: Male 60 T7 T8 (+) Lumbar, fever, night sweats 1: L4 L5 18 Male 57 T11 T12 L1 L2 ()) Lumbar, fever 1: MRI+, magnetic resonance imaging findings suggesting active spinal infection; SAT, standard agglutination test. affected spinal region was performed. ELISA was performed in ten patients and revealed high antibody levels in eight patients. Blood cultures were positive in three patients (16.6%) and all grew B. melitensis; susceptibility tests were not performed. All patients underwent an MRI scan at presentation. The radiological findings [14] were compatible with vertebral osteomyelitis at the lesion site in all patients. Five patients had paravertebral masses and four patients had epidural masses (Table 2). Treatment Duration of therapy varied according to clinical, and radiological response. The median duration of therapy was 48 weeks (range weeks). Three patients received treatment for 24 weeks (Table 3; patients 4, 8 and 17), eight for 48 weeks (Table 3; patients 2, 3, 5, 6, 7, 9, 12 and 18) and six for 72 weeks (Table 3; patients 1, 10, 11, 13, 14 and 15). One patient required treatment discontinuation at approximately 36 weeks because of immune thrombocytopenia (ITP), which was attributed to the antibiotics (Table 3; patient 16). Treatment modifications occurred as a result of adverse effects. Three patients discontinued streptomycin at 2 weeks and two on the first week because of nephrotoxicity (Table 3; patients no 1, 2, 4, 5 and 15). Sulfamethoxazoletrimethoprim was switched to ciprofloxacin because of adverse effects in three patients (Table 3; patients 7, 8 and 13). In six other patients, there were mild side-effects, such as gastrointestinal symptoms, mild elevation of liver function tests or skin reactions. These adverse effects were mild, and did not require treatment modifications. Three of the 18 patients underwent surgical intervention after the initiation of treatment because the spine was mechanically unstable (Table 3; patients 2, 4 and 6). Follow-up and outcome After 24 weeks of treatment, three patients had complete response with resolution of epidural or paravertebral masses on MRI (Table 3, patients 4, 8, 17), despite the fact that two of them continued to have a positive SAT (>1:160). Six patients of the nine with paravertebral or epidural masses on MRI initially, showed only slight slight decrease in soft tissue masses. Fifteen patients with partial remission received a second 24-week treatment cycle. More specifically, these patients continued to experience of lesser severity at the affected spinal area with residual imaging findings. At 48 weeks of treatment, eight had complete remission and discontinued treatment, despite the fact that two patients continued to have a positive SAT (Table 3, patients 2, 3, 5, 6, 7, 9, 12, 18). Resolution of epidural or paravertebral masses occurred in three cases who had at the beginning of treatment. One patient discontinued treatment at 36 weeks because of ITP, as mentioned above (Table 3, patient 16). Six patients with residual imaging findings of vertebral osteomyelitis and residual symptoms received a third treatment cycle for total treatment duration of 72 weeks (Table 3,

5 760 Clinical Microbiology and Infection, Volume 17 Number 5, May 2011 CMI TABLE 3. Treatment and follow-up of the 18 study patients with brucellar spondylitis Patient Treatment (duration in weeks) Findings after 6 months of treatment Findings after 12 months of treatment Findings after 18 months of treatment Findings at the last follow-up Duration of Clinical SAT MRI Clinical SAT MRI Clinical SAT MRI follow-up Clinical SAT MRI 1 Dox + Stm (2); Dox + Rif + T-S (70) Occasional low back 2 Dox + Stm (2); Dox + Rif + T-S (46), surgical treatment Occasional 1: Occasional low back 1: Mild low back 1:2560 ()) 4 years Mild during heavy exercise 1: ()) 1:640 Improved NA NA NA 3 years 5 months 1:2560 ()) ()) 1:640 ()) 3 Dox + Stm (3); Dox + Rif + T-S (45) Neck <1:160 + NA NA NA 5 years 4 Dox + Stm (2); Dox+ Rif + T-S (22), surgical treatment 5 Dox + Stm (1); Dox + Rif + T-S (47) Occasional 6 Dox + Stm (3); Dox + Rif + T-S (45), surgical treatment 7 Dox + Stm (3); Dox + Rif + T-S (45) a Occasional neck NA NA NA NA NA NA 2 years 10 months <1:160 + NA NA NA 8 years Lumbar <1:160 + ()) <1:160 Improved NA NA NA 4 years 10 months <1:160 + NA NA NA 4 years 6 months 8 Dox + Stm (3); Dox + Rif + T-S (21) a ()) 1:1200 ()) NA NA NA NA NA NA 2 years 3 months 9 Dox + Stm (3); Dox + Rif + T-S (45) Low back <1:160 + ()) <1:160 Improved NA NA NA 2 years 1 months 10 Dox + Stm (3); Dox + Rif + T-S (69) Low back <1:160 + Occasional low back 11 Dox + Stm (3); Dox + Rif + T-S (69) Lumbar <1:160 + Occasional 12 Dox + Stm (3) Dox + Rif + T-S (45) Lumbar on effort 13 Dox + Stm (3); Dox + Rif + T-S (69) a Occasional <1:160 + Lumbar during effort <1:160 + Occasional 14 Dox + Stm (3); Dox + Rif + T-S (69) Lumbar <1:160 + Rarely 15 Dox + Stm (1); Dox + Rif + T-S (71) Occasional 16 Dox + Stm (3); Dox + Rif (33) Occasional 1: Rarely <1:160 + ()) <1:160 Improved 3 years 11 months <1:160 + ()) <1:160 Improved 3 years 10 months <1:160 ()) NA NA NA 2 years 1 months <1:160 + Pain on effort <1:160 ()) 2 years 3 months <1: years 3 months ()) 1:1200 ()) ()) <1:160 Improved ()) <1:160 Improved ()) <1:160 Improved Mild during heavy exercise Mild during heavy exercise <1:160 ()) <1:160 ()) 1: ()) 1:2560 ()) 2 months ()) 1:2560 ()) 1: NA NA NA NA NA NA 3 months Mild during weather changes 1:10240 Improved 17 Dox + Rif + T-S (24) ()) 1: ()) NA NA NA NA NA NA 1 months ()) 1:5120 ()) 18 Dox + Stm (3); Dox + Rif + T-S (45) Occasional 1: ()) 1:5120 Improved NA NA NA 4 months ()) 1:5120 Improved Dox, doxycycline; Stm, streptomycin; Rif, rifampicin; T-S, trimethoprim-sulfamethoxazole; MRI+, magnetic resonance imaging findings suggesting active spinal infection; NA, not applicable. a Ciprofloxacin replaced trimethoprim-sulfamethoxazole, which was discontinued after 157 (patient 7), 62 (patient 8) and 120 (patient 13) days of treatment.

6 CMI Ioannou et al. Brucellar spondylodiscitis treatment 761 patients 1, 10, 11, 13, 14, 15). At the end of treatment all patients had complete remission of the disease with no residual paravertebral or epidural masses. However, two patients continued to have a positive SAT. After the completion of treatment, all patients were available for follow-up from 1 up to 96 months (median 36.5 months). During this period, there were no clinical or radiological relapses. Four patients experienced mild sequalae, consisting of mild during heavy exercise or during weather changes. At the last follow-up, seven patients of the thirteen with initially high SAT continued to have positive serology despite clinical and radiological improvement (Table 3). All patients returned to their previous occupation and their everyday activities. Discussion We report the efficacy of prolonged administration (i.e. at least 6 months) of suitable antibiotics in a series of 18 consecutive patients with brucellar spondylitis. The median duration of antimicrobial therapy was 48 weeks (range weeks) and the response rate was 100%. No relapses, mortality or severe sequelae were observed during the long follow-up period. There is no agreement for the conservative treatment of brucellar spondylitis. Issues, such as the optimal antibiotic regimen, the duration of treatment, the role of surgery and the follow-up period remain controversial. A recent review of 8 relevant studies, showed that what matters for the outcome is the duration of treatment, and not the specific recommended regimen used [10]. All our patients were treated for at least 24 weeks, but the majority of them for weeks, because the antibiotics were not discontinued until there was clear improvement of clinical and radiological findings of active brucellosis. This strategy resulted in complete remission, in all our patients. Moreover, at the end of treatment, none had severe functional sequelae, and all returned to their usual occupation. Three (16.66%) of our patients required surgical treatment, after the initiation of antibiotics, for mechanical instability; other large series report surgical intervention at rates ranging from 0% to 30% [4,7 9,17 19]. The major reasons for surgical intervention in the published series are spinal instability and treatment failure. None of our patients required surgery for treatment failure. Relapses are a significant problem of the treatment of brucellar spondylodiscitis because they occur at rates ranging from 0% up to 15% [4,6,7,20,21]. In the present study, none of the 18 patients who completed therapy relapsed during a long follow-up period (1 96 months, median 36.5 months). Relapses usually occur during the first year after infection; most of our patients (14/18, 77.7%) have been followed up for more than 1 year, without relapse. According to a recent position paper, antibiotic regimens suitable for the treatment of brucellosis include combinations of doxycycline with rifampicin for 45 days, and doxycycline for 45 days, plus streptomycin for the initial 2 3 weeks or gentamicin for the initial 7 days [1]. However, an early study has shown that the combination doxycycline/streptomycin may be more effective in brucellar spondylitis [9]. Another study [11] reported that a triple combination may be more successful in treating spondylitis compared to standard regimens. However, these studies do not use uniformly accepted definitions of complete or partial remission, relapse or sequelae, and patients were not treated or followed up homogeneously. Our treatment protocol was strict, the clinical criteria for response were clear, and serial MRI scans were used to make therapeutical decisions. The major disadvantage of protracted administration of an antibiotic regimen is the increased risk for adverse effects. In the present study, eight patients had side-effects requiring treatment modification and one had a severe side effect (i.e. ITP) requiring termination of therapy. The present study has several limitations: there was no control group because of the rarity of the disease; the small number of patients does not allow definite conclusions to be drawn; the response rate of 100% may be attributed to the protracted duration of treatment or to the triple combination of antibiotics, or both; there is no comparison with shorter duration of treatment or another antibiotic regimen; patients had follow-up less than 6 months and in these 4 out of 18 patients is difficult to make a proper outcome assessment. On the other hand, all patients were treated and followed up homogeneously, the definitions were the same for all participating centres, and the follow-up period was sufficient to record possible relapses. In conclusion, prolonged administration of a three-antibiotics combination for at least 6 months seems to comprise an effective treatment strategy for brucellar spondylitis. There is clearly a need for multicentre, prospective therapeutic trials aiming to define the optimal treatment for spondylitis due to Brucella species. Acknowledgements Funding: N. V. Sipsas acknowledges support from the Special Account for Research Funds (ELKE) of the National and Kapodistrian University of Athens, Greece.

7 762 Clinical Microbiology and Infection, Volume 17 Number 5, May 2011 CMI Transparency Declaration N. V. Sipsas has received financial support from the Special Account for Research Funds (ELKE) of the National Kapodistrian University of Athens, Greece. All the other authors: none to declare. References 1. Ariza J, Bosilkovski M, Cascio A et al. Perspectives for the treatment of brucellosis in the 21st century: the Ioannina recommendations. PLoS Med 2007; 4: e Kulowski J, Vinke TH. Undulant (Malta) fever spondylitis: report of a case due to Brucella melitensis, bovine variety, surgically treated. JAMA 1932; 99: Alp E, Doganay M. Current therapeutic strategy in spinal brucellosis. Int J Infect Dis 2008; 12: Alp E, Koc RK, Durak AC et al. Doxycycline plus streptomycin versus ciprofloxacin plus rifampicin in spinal brucellosis. BMC Infect Dis 2006; 6: Colmenero JD, Cisneros JM, Orjuela DL et al. Clinical course and prognosis of Brucella spondylitis. Infection 1992; 20: Colmenero JD, Jiménez-Mejías ME, Sánchez-Lora FJ et al. Pyogenic, tuberculous, and brucellar vertebral osteomyelitis: a descriptive and comparative study of 219 cases. Ann Rheum Dis 1997; 56: Colmenero JD, Ruiz-Mesa JD, Plata A et al. Clinical findings, therapeutic approaches, and outcomes of brucellar vertebral osteomyelitis. Clin Infect Dis 2008; 46: Lifeso RM, Harder E, McCorkell SJ. Spinal brucellosis. J Bone Joint Surg Br 1985; 67: Solera J, Lozano E, Martinez-Alfaro E, Espinosa A, Castillejos ML, Abad L. Brucellar spondylitis: review of 35 cases and literature survey. Clin Infect Dis 1999; 29: Pappas G, Seitaridis N, Tsianos E. Treatment of brucella spondylitis: lesson from an impossible meta-analysis and initial report of efficacy of a fluoroquinolone-containing regimen. Int J Antimicrob Agents 2004; 24: Bayindir Y, Sonmez E, Aladag A, Buyukberber N. Comparison of five antimicrobial regimens for the treatment of brucellar spondylitis: a prospective, randomized study. J Chemother 2003; 15: Ranjbar M, Keramat F, Mamani M et al. Comparison between doxycycline-rifampin-amikacin and doxycycline-rifampin regimens in the treatment of brucellosis. Int J Infect Dis 2007; 11: Young EJ. Brucella species. In: Mandell GL, Bennett JE, Dolin R, eds, Principles and practice of infectious diseases, 6th edn, Vol. 2. Philadelphia, PA: Churchill Livingstone, 2005; Ozaksoy D, Yücesoy K, Yücesoy M, Kovanlikaya I, Yüce A, Naderi S. Brucellar spondylitis: MRI findings. Eur Spine J 2001; 10: Berbari EF, Steckelberg JM, Osman DR: Osteomyelitis. In Mandell, Douglas, and Bennett s Principles and Practice of Infectious Diseases. Volume 1. Sixth edition. Edited by Elsevier, Churchill Livingstone; 2005; Solera J, Rodriguez-Zapata M, Geijo P et al. Doxycycline-rifampin versus doxycycline-streptomycin in treatment of human brucellosis due to Brucella melitensis. Antimicrob Agents Chemother 1995; 39: Belzunegui J, Del Val N, Intxausti JJ, De Dios JR, Queiro R, González C, Rodríguez-Valverde V, Figueroa M. Vertebral osteomyelitis in northern S. Report of 62 cases. Clin Exp Rheumatol. 1999; 17: Haddad FS, Fahl H, Haddad SF. Brucellar spondylitis. Review of 21 cases. J Med Liban. 1987; 37: Taşova Y, Saltoğlu N, Sahin G, Aksu HS. Osteoarthricular involvement of brucellosis in Turkey. Clin Rheumatol. 1999; 18: Solera J, Martinez-Alfaro E, Espinosa A, Castillejos ML, Geijo P, Rodriguez-Zapata M. Multivariate model for predicting relapse in human brucellosis. J Infect 1998; 36: Tekkök IH, Berker M, Ozcan OE, Ozgen T, Akalin E. Brucellosis of the spine. Neurosurgery. 1993; 33: