- Email: [email protected]
Postoperative spondylodiscitis caused by Mycobacterium bovis BCG: a case study
The Spine Journal 12 (2012) e1–e5
Case Report
Postoperative spondylodiscitis caused by Mycobacterium bovis BCG: a case study T~ onu R€atsep, MD, PhD* Department of Neurology and Neurosurgery, University of Tartu, L. Puusepa 8, 51014 Tartu, Estonia Received 12 November 2011; revised 17 April 2012; accepted 13 October 2012
Abstract
BACKGROUND CONTEXT: Postoperative spondylodiscitis (PSD) is a known complication of lumbar disc surgery. The etiology of the disease is usually bacterial, but several uncommon infectious agents have also been described; however, there are no reports about postoperative colonization with Mycobacterium bovis bacille Calmette-Guerin after lumbar discectomy. PURPOSE: To describe the case of PSD caused by M. bovis BCG, and to discuss diagnostic and therapeutic interventions as well as possible pathogenic mechanisms of the disease. STUDY DESIGN: Case report and review of the literature. METHODS: A 31-year-old man was operated on because of L4–L5 lumbar disc herniation. Two months later, the patient presented with gradual increase of back pain, and magnetic resonance imaging confirmed PSD. He started to receive antibacterial treatment and was reoperated on because of progressive neurological deficits due to epidural abscess. Neurological status improved, but a fistula developed with intermittent pus drainage from the operative scar. RESULTS: Microbiological cultures were repeatedly obtained from the pus, but all the initial stains and cultures were negative. Four months after the reoperation, the culture for M. bovis BCG from the pus appeared to be positive. The patient received antituberculosis regimen, including isoniazid, rifampin, ethambutol, and ofloxacin. The clinical symptoms resolved, and antituberculosis treatment was discontinued after 14 months. CONCLUSIONS: Mycobacterium bovis BCG must be considered in the differential diagnosis of PSD. Microbiological analysis and radiological studies are vital components in diagnosis; if there is any suspicion of BCG osteomyelitis, proper diagnostic and therapeutic management must be instituted without delay to avoid an unfavorable outcome. Ó 2012 Elsevier Inc. All rights reserved.
Keywords:
Postoperative spondylodiscitis; Mycobacterium bovis BCG; Tuberculosis
Introduction Postoperative spondylodiscitis (PSD) is a known complication of lumbar disc surgery. Although the exact incidence of PSD is difficult to ascertain, it ranges from 0.6% to 5% [1], and the absolute number of cases might be increasing because of the increasing number of spine surgeries performed. Factors such as malnutrition, diabetes, substance abuse, malignancy, chronic comorbidities, obesity, immunocompromised states, and excessive instrumentation can increase the development of PSD. The etiology
of PSD is commonly bacterial, but in recent years, several uncommon infectious agents have been described including cases caused by fungi or mycobacteria [2–4]. The purpose of this case report was to present an atypical PSD associated with epidural abscess, induced by bacille Calmette-Guerin (BCG), and to discuss diagnostic and therapeutic interventions as well as the possible pathogenic mechanisms of the disease.
Case report FDA device/drug status: Not applicable. Author disclosures: TR: Nothing to disclose. * Corresponding author. Department of Neurology and Neurosurgery, University of Tartu, L. Puusepa 8, 51014 Tartu, Estonia. Tel.: (372) 7318-542; fax: (372) 7-318-509. E-mail address: [email protected] (T. R€atsep) 1529-9430/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.spinee.2012.10.024
A 31-year-old man presented with intermittent low back pain combined with radicular pain because of L4–L5 lumbar disc herniation. Medical history of the patient was negative for any background infections, substance abuse, diabetes, tuberculosis, or immunocompromised states.
e2
T. R€atsep / The Spine Journal 12 (2012) e1–e5
Fig. 1. Preoperative sagittal T2-weighted magnetic resonance imaging scan demonstrates L4–L5 disc herniation without evidence of indolent infection before the surgery.
The results of the routine laboratory examinations were normal. The diagnosis was confirmed with magnetic resonance imaging (MRI) (Fig. 1), and the patient was operated on with standard posterior microdiscectomy approach. Postoperatively, his preoperative symptoms improved, and the patient returned home in good conditions. Two months later, the patient presented with gradual increase of back pain. He was afebrile, with normal motor and neurologic functions. The MRI confirmed destructive changes in the L4–L5 disc space and the neighboring end plates in favor of an infective process. His routine cell counts were normal, but C-reactive protein (CRP) was 29 mg/L (!5 mg/L). Peripheral blood cultures showed no
growth. Thorough investigation for possible urinary tract or other coinfections was negative. Testing for antibody to human immunodeficiency virus by enzyme-linked immunosorbent assay was negative twice. He received the diagnosis of PSD, and antibacterial treatment was empirically started with cefazolin followed by intravenous oxacillin to cover Staphylococcus as the most likely bacterial pathogen. Despite the treatment, signs of radicular compression developed including irradiating pain and decreased sensation in his leg. The following MRI demonstrated an epidural abscess at the operated level with anterior dural sac compression (Fig. 2). The value of CRP showed moderate increase (32 mg/L). Because of the progressive neurological deficits and MRI findings, surgical intervention was decided and the patient was reoperated on. The epidural abscess was evacuated with posterior approach, and aerobic, anaerobic, and fungal cultures were obtained, but all microbiological stains and cultures from the pus were negative. Neurological status improved immediately after the surgical treatment. He continued to receive antibiotics (oxacillin had to be replaced by clindamycin because of allergic leucopenic reaction), and his CRP value decreased (14 mg/L). Two months after the reoperation, a fistula developed with intermittent pus drainage from his operative scar in the lumbar region. He continued to need analgesics because of low back pain, and CRP value elevated again to 29 mg/L. Aerobic, anaerobic, and fungal cultures were repeatedly obtained from the pus, but all microbiological stains and cultures were negative. However, 4 months after the reoperation, the culture for Mycobacterium bovis BCG from the fistula appeared to be positive according to GenoType Mycobacterium tuberculosis complex (MTBC) (Hain Lifescience GmbH, Nehren, Germany). The patient denied any history of a contact with tuberculosis patients in the past, but he had received immunization with BCG vaccine twice, first at birth and second during the adolescence period. The
Fig. 2. Postoperative (Left) sagittal and (Right) axial T1-weighted magnetic resonance images with contrast enhancement show destruction of the L4 and L5 vertebral end plates, involvement of the intervertebral disc space, and an epidural abscess with anterior dural sac compression.
T. R€atsep / The Spine Journal 12 (2012) e1–e5
e3
Fig. 3. (Left) Sagittal and (Right) axial T1-weighted magnetic resonance images with contrast enhancement at 4 months after antituberculosis treatment reveal increased signal and formation of bone fusion within the involved vertebrae together with paravertebral mass at the anterior aspect of the vertebral bodies and left psoas muscle with signal abnormalities suggesting of an abscess.
isolated strain was resistant to pyrazinamide and sensitive to all other first- and second-line antituberculosis drugs. Based on the aforementioned findings, the patient started to receive an antituberculosis regimen, including isoniazid, rifampin, ethambutol, and ofloxacin. The positive culture for M. bovis BCG was repeated for four times during the first month of the treatment and turned negative after that. The hospital course was prolonged because of difficulty with pain control and immobility due to that. Four months after the beginning of the antituberculosis treatment, paravertebral abscess formation was described on MRI (Fig. 3), but by that time, the patient’s general conditions had already started to improve, and no surgical evacuation of the abscess was undertaken. Pus drainage from the fistula diminished gradually, and by the end of the 10th month of the treatment, it was totally abolished. The MRI at 1 year after the antituberculosis treatment revealed regression of the paravertebral abscess without the development of significant spinal deformity or lumbar stenosis (Fig. 4). Most of the clinical symptoms of the patient resolved, and antituberculosis treatment was discontinued after 14 months. He
continues to have occasional low back pain but was able to stop the utilization of analgetics and has no clinical or imaging findings of spinal instability or recurrent infection.
Discussion BCG vaccine is a live attenuated derivative of the virulent strain of M. bovis that was first used for immunization against tuberculosis in 1921. As about 100 million children receive the vaccine annually, it is one of the most widely used human vaccines and remains to be one of the safest vaccines available [5]. However, cumulative data from decades of research have revealed rare but serious complications with the vaccine [6,7]. Localized abscesses at the site of inoculation, regional lymphadenopathy, musculoskeletal complications, including arthralgias, reactive arthritis, and osteoarticular infections, and disseminated disease in immunocompromised hosts, are uncommon but well-recognized complications after vaccination and intravesical BCG therapy [6–9]. The incidence and type of
Fig. 4. (Left) Sagittal and (Right) axial T1-weighted magnetic resonance images with contrast enhancement obtained at 10 months after the initiation of antituberculosis treatment show regression of the paravertebral abscess, postinfectious L4–L5 vertebral fusion, and no spinal canal stenosis in favor of successful treatment.
e4
T. R€atsep / The Spine Journal 12 (2012) e1–e5
complications associated with BCG vaccination is variable and could depend on the particular strain, dose, and route of administration. The risk of osteomyelitis after BCG immunization has been evaluated already decades ago. The highest incidence of osteitis caused by BCG vaccination, that is, one case per 3,000 vaccinated newborns was observed in 1973 in Sweden. B€ ottiger et al. [10] reported that of the 152 patients, only in six, the spine, which is the common site of osteitis caused by tuberculosis, was affected. However, extremely long incubation periods were observed, that is, 6 years in two children and 12 years in one. In 1995, Kr€ oger et al. [8] evaluated the frequency, clinical features, and prognosis of patients with osteitis caused by BCG vaccination in Finland according to national database from 1960 to 1988. During the study period, three different BCG vaccine preparations were used, and diagnostic criteria of BCG osteitis were fulfilled in 222 children. The researchers reported that the frequency of BCG osteitis, but not the clinical parameters, was closely associated with the vaccine preparation used. Moreover, since 1992 [11], several cases of vertebral osteomyelitis caused by M. bovis BCG have been described after the intravesical BCG immunotherapy for bladder cancer [9,12–14]. Although the exact cause of postprocedural discitis has always been controversial, the possible sources of infection are direct inoculation at the time of surgery, soiling of the incision in the acute postoperative phase, or dissemination through hematogenous seeding. Direct inoculation is the most probable cause if the offending etiologic agent is Staphylococcus aureus or other Staphylococcus species; however, several cases of postoperative Pott disease caused by M. tuberculosis have been presented after lumbar discectomy, and the possibility of a direct infection during the surgery has also been discussed as the pathogenetic mechanism of the disease [2–4,15]. Still, hematogenous spread from indolent primary foci of the disease was considered as the most likely cause of the infection. Diagnostic difficulties were evident in all patients, initial stains and cultures were usually negative for any bacterial infections, and empiric antibiotics were started in most patients. As a rule, M. tuberculosis was identified as the etiologic agent, and antituberculosis regimen was started only after considerable time delay. The occurrence of nontuberculous mycobacteria infections has been reported in surgical patients not only in a wide variety of settings but also in the field of spinal surgery [16,17]. The use of colonized aqueous solutions and inadequate sterilization or disinfection of surgical equipment are often factors in these infections [18]. In 2001, Astagneau et al. [19] described 58 cases of postoperative spinal infections caused by Mycobacterium xenopi contamination during discovertebral surgery. Nontuberculous mycobacterial infections are more associated with immunocompromised patients but can occur in any postoperative wound and must be considered if wounds show delayed healing or insidious onset of the infection, have
sterile cultures, and demonstrate lack of response to antibiotics used for acute pyogenic infections. There are no references to other cases of PSD caused by M. bovis BCG in the literature, and the rarity of the organism would strongly argue against the possibility of direct contamination as the cause of the disease. Overall, there has been an ongoing dispute about the underlying mechanisms of systemic complications related to BCG therapy or vaccination. It appears that both direct infectious processes and hypersensitivity reactions can contribute to the development of complications, and the higher incidence has been ascribed to the injection technique, route of delivery, immune status of the host, dose, and strain that has been used [5,7,8,20]. However, the finding of a positive culture of BCG from a distant site is a direct indication of dissemination, and hematogenous spread is the leading pathogenetic mechanism. There is substantial epidemiological evidence that genetic factors in part determine the susceptibility to mycobacteria, and specific gene defects have been identified impairing the host’s defense against mycobacterial infections [21–23]. Considering the rarity of the case, genetic differences in immune responses may also have affected susceptibility to the mycobacterial infection in our patient. It is very important to distinguish M. bovis BCG strains from the MTBC, which includes M. tuberculosis, M. bovis, M. bovis BCG, Mycobacterium africanum, Mycobacterium microti, and Mycobacterium canetti. Routine laboratory tests do not give substantial information about the diagnosis, although only moderately elevated CRP values are more characteristic to nonpyogenic infections. Our diagnosis was based on a commercially available DNA strip assay (GenoType MTBC; Hain Lifescience GmbH, Nehren, Germany) intended for the differentiation of members of the MTBC and identification of M. bovis BCG. This assay is based on gyrB DNA sequence polymorphisms and the RD1 deletion of M. bovis BCG. Conventional differentiation of the bacteria has relied on the growth characteristics and biochemical properties of the strains. Most assays have required ample culture material, usually grown on solid slants for weeks; special media have been necessary; and sometimes borderline results have been obtained. The GenoType MTBC assay has been tested and found to allow reliable and quick differentiation of MTBC species [24]. There is no universally accepted treatment protocol neither for PSD nor for the complications caused by M. bovis BCG. The cornerstones of nonoperative management of PSD are immobilization/bracing and prolonged organismspecific antibiotic therapy [25]. As all strains of M. bovis are intrinsically resistant to pyrazinamide, most commonly used drugs for the treatment of disseminated BCG complications are isoniazid, rifampin, ethambutol, and streptomycin [7]. Although treatment depends on the nature of the complication and its severity, in case of disseminated systemic infection or specific organ involvement, as in the present case, at least triple-regimen antituberculosis therapy is warranted [20]. In cases of aggressive PSD, surgical
T. R€atsep / The Spine Journal 12 (2012) e1–e5
intervention may be necessary for abscess drainage and occasionally, perhaps, for spine stabilization. Although there is no information about any other similar cases and exact pathomechanisms of the disease are unclear, it has to be concluded that in previously vaccinated patients M. bovis BCG must be considered in the differential diagnosis of PSD. When a case of PSD does not follow the expected course and if there is any suspicion of BCG osteomyelitis, proper diagnostic and therapeutic management must be instituted without delay to avoid an unfavorable outcome. References [1] Chaudhary SB, Vives MJ, Basra SK, Reiter MF. Postoperative spinal wound infections and postprocedural diskitis. J Spinal Cord Med 2007;30:441–51. [2] Lotfinia I, Vahedi P. Late-onset post-diskectomy tuberculosis at the same operated lumbar level: case report and review of literature. Eur Spine J 2010;19(Suppl 2):S226–32. [3] Jeon do W, Chang BS, Jeung UO, et al. A case of postoperative tuberculous spondylitis with a bizarre course. Clin Orthop Surg 2009;1:58–62. [4] Celik SE. Pott disease mimics postsurgical pyogenic spondylodiscitis. J Neurosurg Sci 2009;53:161–4. [5] Murphy D, Corner LA, Gormley E. Adverse reactions to Mycobacterium bovis bacille Calmette-Guerin (BCG) vaccination against tuberculosis in humans, veterinary animals and wildlife species. Tuberculosis (Edinb) 2008;88:344–57. [6] Lotte A, Wasz-H€ ockert O, Poisson N, et al. BCG complications. Estimates of the risks among vaccinated subjects and statistical analysis of their main characteristics. Adv Tuberc Res 1984;21:107–93. [7] Talbot EA, Perkins MD, Silva SF, Frothingham R. Disseminated bacille Calmette-Guerin disease after vaccination: case report and review. Clin Infect Dis 1997;24:1139–46. [8] Kr€ oger L, Korppi M, Brander E, et al. Osteitis caused by bacille Calmette-Guerin vaccination: a retrospective analysis of 222 cases. J Infect Dis 1995;172:574–6. [9] Josephson CB, Al-Azri S, Smyth DJ, et al. A case of Pott’s disease with epidural abscess and probable cerebral tuberculoma following Bacillus Calmette-Guerin therapy for superficial bladder cancer. Can J Infect Dis Med Microbiol 2010;21:e75–8.
e5
[10] B€ottiger M, Romanus V, de Verdier C, Boman G. Osteitis and other complications caused by generalized BCG-itis. Experiences in Sweden. Acta Paediatr Scand 1982;71:471–8. [11] Katz DS, Wogalter H, D’Esposito RF, Cunha BA. Mycobacterium bovis vertebral osteomyelitis and psoas abscess after intravesical BCG therapy for bladder carcinoma. Urology 1992;40:63–6. [12] Aljada IS, Crane JK, Corriere N, et al. Mycobacterium bovis BCG causing vertebral osteomyelitis (Pott’s disease) following intravesical BCG therapy. J Clin Microbiol 1999;37:2106–8. [13] Abu-Nader R, Terrell CL. Mycobacterium bovis vertebral osteomyelitis as a complication of intravesical BCG use. Mayo Clin Proc 2002;77:393–7. [14] Mavrogenis AF, Sakellariou VI, Tsiodras S, Papagelopoulos PJ. Late Mycobacterium bovis spondylitis after intravesical BCG therapy. Joint Bone Spine 2009;76:296–300. [15] Kaplan ES. Post-discectomy tuberculosis abscess. Case report. J Neurosurg 1973;38:358–61. [16] Kim SH, Son DW, Lee SW, Song GS. An unusual case of postoperative spondylitis caused by mycobacterium intracellulare in an immunosuppressed patient. J Korean Neurosurg Soc 2011;50: 460–3. [17] Zimba FA. Mycobacterium smegmatis infection of a lumbar spine instrumented fusion. Am J Infect Dis 2011;7:24–7. [18] Phillips MS, von Reyn CF. Nosocomial infections due to nontuberculous mycobacteria. Clin Infect Dis 2001;33:1363–74. [19] Astagneau P, Desplaces N, Vincent V, et al. Mycobacterium xenopi spinal infections after discovertebral surgery: investigation and screening of a large outbreak. Lancet 2001;358:747–51. [20] Grange JM. Complications of bacille Calmette-Guerin (BCG) vaccination and immunotherapy and their management. Commun Dis Public Health 1998;1:84–8. [21] Newport MJ, Huxley CM, Huston S, et al. A mutation in the interferon-gamma-receptor gene and susceptibility to mycobacterial infection. N Engl J Med 1996;335:1941–9. [22] Altare F, Durandy A, Lammas D, et al. Impairment of mycobacterial immunity in human interleukin-12 receptor deficiency. Science 1998;280:1432–5. [23] Newport MJ, Nejentsev S. Genetics of susceptibility to tuberculosis in humans. Monaldi Arch Chest Dis 2004;61:102–11. [24] Richter E, Weizenegger M, R€usch-Gerdes S, Niemann S. Evaluation of genotype MTBC assay for differentiation of clinical Mycobacterium tuberculosis complex isolates. J Clin Microbiol 2003;41: 2672–5. [25] Silber JS, Anderson DG, Vaccaro AR, et al, NASS. Management of postprocedural discitis. Spine J 2002;2:279–87.
Case Report
Postoperative spondylodiscitis caused by Mycobacterium bovis BCG: a case study T~ onu R€atsep, MD, PhD* Department of Neurology and Neurosurgery, University of Tartu, L. Puusepa 8, 51014 Tartu, Estonia Received 12 November 2011; revised 17 April 2012; accepted 13 October 2012
Abstract
BACKGROUND CONTEXT: Postoperative spondylodiscitis (PSD) is a known complication of lumbar disc surgery. The etiology of the disease is usually bacterial, but several uncommon infectious agents have also been described; however, there are no reports about postoperative colonization with Mycobacterium bovis bacille Calmette-Guerin after lumbar discectomy. PURPOSE: To describe the case of PSD caused by M. bovis BCG, and to discuss diagnostic and therapeutic interventions as well as possible pathogenic mechanisms of the disease. STUDY DESIGN: Case report and review of the literature. METHODS: A 31-year-old man was operated on because of L4–L5 lumbar disc herniation. Two months later, the patient presented with gradual increase of back pain, and magnetic resonance imaging confirmed PSD. He started to receive antibacterial treatment and was reoperated on because of progressive neurological deficits due to epidural abscess. Neurological status improved, but a fistula developed with intermittent pus drainage from the operative scar. RESULTS: Microbiological cultures were repeatedly obtained from the pus, but all the initial stains and cultures were negative. Four months after the reoperation, the culture for M. bovis BCG from the pus appeared to be positive. The patient received antituberculosis regimen, including isoniazid, rifampin, ethambutol, and ofloxacin. The clinical symptoms resolved, and antituberculosis treatment was discontinued after 14 months. CONCLUSIONS: Mycobacterium bovis BCG must be considered in the differential diagnosis of PSD. Microbiological analysis and radiological studies are vital components in diagnosis; if there is any suspicion of BCG osteomyelitis, proper diagnostic and therapeutic management must be instituted without delay to avoid an unfavorable outcome. Ó 2012 Elsevier Inc. All rights reserved.
Keywords:
Postoperative spondylodiscitis; Mycobacterium bovis BCG; Tuberculosis
Introduction Postoperative spondylodiscitis (PSD) is a known complication of lumbar disc surgery. Although the exact incidence of PSD is difficult to ascertain, it ranges from 0.6% to 5% [1], and the absolute number of cases might be increasing because of the increasing number of spine surgeries performed. Factors such as malnutrition, diabetes, substance abuse, malignancy, chronic comorbidities, obesity, immunocompromised states, and excessive instrumentation can increase the development of PSD. The etiology
of PSD is commonly bacterial, but in recent years, several uncommon infectious agents have been described including cases caused by fungi or mycobacteria [2–4]. The purpose of this case report was to present an atypical PSD associated with epidural abscess, induced by bacille Calmette-Guerin (BCG), and to discuss diagnostic and therapeutic interventions as well as the possible pathogenic mechanisms of the disease.
Case report FDA device/drug status: Not applicable. Author disclosures: TR: Nothing to disclose. * Corresponding author. Department of Neurology and Neurosurgery, University of Tartu, L. Puusepa 8, 51014 Tartu, Estonia. Tel.: (372) 7318-542; fax: (372) 7-318-509. E-mail address: [email protected] (T. R€atsep) 1529-9430/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.spinee.2012.10.024
A 31-year-old man presented with intermittent low back pain combined with radicular pain because of L4–L5 lumbar disc herniation. Medical history of the patient was negative for any background infections, substance abuse, diabetes, tuberculosis, or immunocompromised states.
e2
T. R€atsep / The Spine Journal 12 (2012) e1–e5
Fig. 1. Preoperative sagittal T2-weighted magnetic resonance imaging scan demonstrates L4–L5 disc herniation without evidence of indolent infection before the surgery.
The results of the routine laboratory examinations were normal. The diagnosis was confirmed with magnetic resonance imaging (MRI) (Fig. 1), and the patient was operated on with standard posterior microdiscectomy approach. Postoperatively, his preoperative symptoms improved, and the patient returned home in good conditions. Two months later, the patient presented with gradual increase of back pain. He was afebrile, with normal motor and neurologic functions. The MRI confirmed destructive changes in the L4–L5 disc space and the neighboring end plates in favor of an infective process. His routine cell counts were normal, but C-reactive protein (CRP) was 29 mg/L (!5 mg/L). Peripheral blood cultures showed no
growth. Thorough investigation for possible urinary tract or other coinfections was negative. Testing for antibody to human immunodeficiency virus by enzyme-linked immunosorbent assay was negative twice. He received the diagnosis of PSD, and antibacterial treatment was empirically started with cefazolin followed by intravenous oxacillin to cover Staphylococcus as the most likely bacterial pathogen. Despite the treatment, signs of radicular compression developed including irradiating pain and decreased sensation in his leg. The following MRI demonstrated an epidural abscess at the operated level with anterior dural sac compression (Fig. 2). The value of CRP showed moderate increase (32 mg/L). Because of the progressive neurological deficits and MRI findings, surgical intervention was decided and the patient was reoperated on. The epidural abscess was evacuated with posterior approach, and aerobic, anaerobic, and fungal cultures were obtained, but all microbiological stains and cultures from the pus were negative. Neurological status improved immediately after the surgical treatment. He continued to receive antibiotics (oxacillin had to be replaced by clindamycin because of allergic leucopenic reaction), and his CRP value decreased (14 mg/L). Two months after the reoperation, a fistula developed with intermittent pus drainage from his operative scar in the lumbar region. He continued to need analgesics because of low back pain, and CRP value elevated again to 29 mg/L. Aerobic, anaerobic, and fungal cultures were repeatedly obtained from the pus, but all microbiological stains and cultures were negative. However, 4 months after the reoperation, the culture for Mycobacterium bovis BCG from the fistula appeared to be positive according to GenoType Mycobacterium tuberculosis complex (MTBC) (Hain Lifescience GmbH, Nehren, Germany). The patient denied any history of a contact with tuberculosis patients in the past, but he had received immunization with BCG vaccine twice, first at birth and second during the adolescence period. The
Fig. 2. Postoperative (Left) sagittal and (Right) axial T1-weighted magnetic resonance images with contrast enhancement show destruction of the L4 and L5 vertebral end plates, involvement of the intervertebral disc space, and an epidural abscess with anterior dural sac compression.
T. R€atsep / The Spine Journal 12 (2012) e1–e5
e3
Fig. 3. (Left) Sagittal and (Right) axial T1-weighted magnetic resonance images with contrast enhancement at 4 months after antituberculosis treatment reveal increased signal and formation of bone fusion within the involved vertebrae together with paravertebral mass at the anterior aspect of the vertebral bodies and left psoas muscle with signal abnormalities suggesting of an abscess.
isolated strain was resistant to pyrazinamide and sensitive to all other first- and second-line antituberculosis drugs. Based on the aforementioned findings, the patient started to receive an antituberculosis regimen, including isoniazid, rifampin, ethambutol, and ofloxacin. The positive culture for M. bovis BCG was repeated for four times during the first month of the treatment and turned negative after that. The hospital course was prolonged because of difficulty with pain control and immobility due to that. Four months after the beginning of the antituberculosis treatment, paravertebral abscess formation was described on MRI (Fig. 3), but by that time, the patient’s general conditions had already started to improve, and no surgical evacuation of the abscess was undertaken. Pus drainage from the fistula diminished gradually, and by the end of the 10th month of the treatment, it was totally abolished. The MRI at 1 year after the antituberculosis treatment revealed regression of the paravertebral abscess without the development of significant spinal deformity or lumbar stenosis (Fig. 4). Most of the clinical symptoms of the patient resolved, and antituberculosis treatment was discontinued after 14 months. He
continues to have occasional low back pain but was able to stop the utilization of analgetics and has no clinical or imaging findings of spinal instability or recurrent infection.
Discussion BCG vaccine is a live attenuated derivative of the virulent strain of M. bovis that was first used for immunization against tuberculosis in 1921. As about 100 million children receive the vaccine annually, it is one of the most widely used human vaccines and remains to be one of the safest vaccines available [5]. However, cumulative data from decades of research have revealed rare but serious complications with the vaccine [6,7]. Localized abscesses at the site of inoculation, regional lymphadenopathy, musculoskeletal complications, including arthralgias, reactive arthritis, and osteoarticular infections, and disseminated disease in immunocompromised hosts, are uncommon but well-recognized complications after vaccination and intravesical BCG therapy [6–9]. The incidence and type of
Fig. 4. (Left) Sagittal and (Right) axial T1-weighted magnetic resonance images with contrast enhancement obtained at 10 months after the initiation of antituberculosis treatment show regression of the paravertebral abscess, postinfectious L4–L5 vertebral fusion, and no spinal canal stenosis in favor of successful treatment.
e4
T. R€atsep / The Spine Journal 12 (2012) e1–e5
complications associated with BCG vaccination is variable and could depend on the particular strain, dose, and route of administration. The risk of osteomyelitis after BCG immunization has been evaluated already decades ago. The highest incidence of osteitis caused by BCG vaccination, that is, one case per 3,000 vaccinated newborns was observed in 1973 in Sweden. B€ ottiger et al. [10] reported that of the 152 patients, only in six, the spine, which is the common site of osteitis caused by tuberculosis, was affected. However, extremely long incubation periods were observed, that is, 6 years in two children and 12 years in one. In 1995, Kr€ oger et al. [8] evaluated the frequency, clinical features, and prognosis of patients with osteitis caused by BCG vaccination in Finland according to national database from 1960 to 1988. During the study period, three different BCG vaccine preparations were used, and diagnostic criteria of BCG osteitis were fulfilled in 222 children. The researchers reported that the frequency of BCG osteitis, but not the clinical parameters, was closely associated with the vaccine preparation used. Moreover, since 1992 [11], several cases of vertebral osteomyelitis caused by M. bovis BCG have been described after the intravesical BCG immunotherapy for bladder cancer [9,12–14]. Although the exact cause of postprocedural discitis has always been controversial, the possible sources of infection are direct inoculation at the time of surgery, soiling of the incision in the acute postoperative phase, or dissemination through hematogenous seeding. Direct inoculation is the most probable cause if the offending etiologic agent is Staphylococcus aureus or other Staphylococcus species; however, several cases of postoperative Pott disease caused by M. tuberculosis have been presented after lumbar discectomy, and the possibility of a direct infection during the surgery has also been discussed as the pathogenetic mechanism of the disease [2–4,15]. Still, hematogenous spread from indolent primary foci of the disease was considered as the most likely cause of the infection. Diagnostic difficulties were evident in all patients, initial stains and cultures were usually negative for any bacterial infections, and empiric antibiotics were started in most patients. As a rule, M. tuberculosis was identified as the etiologic agent, and antituberculosis regimen was started only after considerable time delay. The occurrence of nontuberculous mycobacteria infections has been reported in surgical patients not only in a wide variety of settings but also in the field of spinal surgery [16,17]. The use of colonized aqueous solutions and inadequate sterilization or disinfection of surgical equipment are often factors in these infections [18]. In 2001, Astagneau et al. [19] described 58 cases of postoperative spinal infections caused by Mycobacterium xenopi contamination during discovertebral surgery. Nontuberculous mycobacterial infections are more associated with immunocompromised patients but can occur in any postoperative wound and must be considered if wounds show delayed healing or insidious onset of the infection, have
sterile cultures, and demonstrate lack of response to antibiotics used for acute pyogenic infections. There are no references to other cases of PSD caused by M. bovis BCG in the literature, and the rarity of the organism would strongly argue against the possibility of direct contamination as the cause of the disease. Overall, there has been an ongoing dispute about the underlying mechanisms of systemic complications related to BCG therapy or vaccination. It appears that both direct infectious processes and hypersensitivity reactions can contribute to the development of complications, and the higher incidence has been ascribed to the injection technique, route of delivery, immune status of the host, dose, and strain that has been used [5,7,8,20]. However, the finding of a positive culture of BCG from a distant site is a direct indication of dissemination, and hematogenous spread is the leading pathogenetic mechanism. There is substantial epidemiological evidence that genetic factors in part determine the susceptibility to mycobacteria, and specific gene defects have been identified impairing the host’s defense against mycobacterial infections [21–23]. Considering the rarity of the case, genetic differences in immune responses may also have affected susceptibility to the mycobacterial infection in our patient. It is very important to distinguish M. bovis BCG strains from the MTBC, which includes M. tuberculosis, M. bovis, M. bovis BCG, Mycobacterium africanum, Mycobacterium microti, and Mycobacterium canetti. Routine laboratory tests do not give substantial information about the diagnosis, although only moderately elevated CRP values are more characteristic to nonpyogenic infections. Our diagnosis was based on a commercially available DNA strip assay (GenoType MTBC; Hain Lifescience GmbH, Nehren, Germany) intended for the differentiation of members of the MTBC and identification of M. bovis BCG. This assay is based on gyrB DNA sequence polymorphisms and the RD1 deletion of M. bovis BCG. Conventional differentiation of the bacteria has relied on the growth characteristics and biochemical properties of the strains. Most assays have required ample culture material, usually grown on solid slants for weeks; special media have been necessary; and sometimes borderline results have been obtained. The GenoType MTBC assay has been tested and found to allow reliable and quick differentiation of MTBC species [24]. There is no universally accepted treatment protocol neither for PSD nor for the complications caused by M. bovis BCG. The cornerstones of nonoperative management of PSD are immobilization/bracing and prolonged organismspecific antibiotic therapy [25]. As all strains of M. bovis are intrinsically resistant to pyrazinamide, most commonly used drugs for the treatment of disseminated BCG complications are isoniazid, rifampin, ethambutol, and streptomycin [7]. Although treatment depends on the nature of the complication and its severity, in case of disseminated systemic infection or specific organ involvement, as in the present case, at least triple-regimen antituberculosis therapy is warranted [20]. In cases of aggressive PSD, surgical
T. R€atsep / The Spine Journal 12 (2012) e1–e5
intervention may be necessary for abscess drainage and occasionally, perhaps, for spine stabilization. Although there is no information about any other similar cases and exact pathomechanisms of the disease are unclear, it has to be concluded that in previously vaccinated patients M. bovis BCG must be considered in the differential diagnosis of PSD. When a case of PSD does not follow the expected course and if there is any suspicion of BCG osteomyelitis, proper diagnostic and therapeutic management must be instituted without delay to avoid an unfavorable outcome. References [1] Chaudhary SB, Vives MJ, Basra SK, Reiter MF. Postoperative spinal wound infections and postprocedural diskitis. J Spinal Cord Med 2007;30:441–51. [2] Lotfinia I, Vahedi P. Late-onset post-diskectomy tuberculosis at the same operated lumbar level: case report and review of literature. Eur Spine J 2010;19(Suppl 2):S226–32. [3] Jeon do W, Chang BS, Jeung UO, et al. A case of postoperative tuberculous spondylitis with a bizarre course. Clin Orthop Surg 2009;1:58–62. [4] Celik SE. Pott disease mimics postsurgical pyogenic spondylodiscitis. J Neurosurg Sci 2009;53:161–4. [5] Murphy D, Corner LA, Gormley E. Adverse reactions to Mycobacterium bovis bacille Calmette-Guerin (BCG) vaccination against tuberculosis in humans, veterinary animals and wildlife species. Tuberculosis (Edinb) 2008;88:344–57. [6] Lotte A, Wasz-H€ ockert O, Poisson N, et al. BCG complications. Estimates of the risks among vaccinated subjects and statistical analysis of their main characteristics. Adv Tuberc Res 1984;21:107–93. [7] Talbot EA, Perkins MD, Silva SF, Frothingham R. Disseminated bacille Calmette-Guerin disease after vaccination: case report and review. Clin Infect Dis 1997;24:1139–46. [8] Kr€ oger L, Korppi M, Brander E, et al. Osteitis caused by bacille Calmette-Guerin vaccination: a retrospective analysis of 222 cases. J Infect Dis 1995;172:574–6. [9] Josephson CB, Al-Azri S, Smyth DJ, et al. A case of Pott’s disease with epidural abscess and probable cerebral tuberculoma following Bacillus Calmette-Guerin therapy for superficial bladder cancer. Can J Infect Dis Med Microbiol 2010;21:e75–8.
e5
[10] B€ottiger M, Romanus V, de Verdier C, Boman G. Osteitis and other complications caused by generalized BCG-itis. Experiences in Sweden. Acta Paediatr Scand 1982;71:471–8. [11] Katz DS, Wogalter H, D’Esposito RF, Cunha BA. Mycobacterium bovis vertebral osteomyelitis and psoas abscess after intravesical BCG therapy for bladder carcinoma. Urology 1992;40:63–6. [12] Aljada IS, Crane JK, Corriere N, et al. Mycobacterium bovis BCG causing vertebral osteomyelitis (Pott’s disease) following intravesical BCG therapy. J Clin Microbiol 1999;37:2106–8. [13] Abu-Nader R, Terrell CL. Mycobacterium bovis vertebral osteomyelitis as a complication of intravesical BCG use. Mayo Clin Proc 2002;77:393–7. [14] Mavrogenis AF, Sakellariou VI, Tsiodras S, Papagelopoulos PJ. Late Mycobacterium bovis spondylitis after intravesical BCG therapy. Joint Bone Spine 2009;76:296–300. [15] Kaplan ES. Post-discectomy tuberculosis abscess. Case report. J Neurosurg 1973;38:358–61. [16] Kim SH, Son DW, Lee SW, Song GS. An unusual case of postoperative spondylitis caused by mycobacterium intracellulare in an immunosuppressed patient. J Korean Neurosurg Soc 2011;50: 460–3. [17] Zimba FA. Mycobacterium smegmatis infection of a lumbar spine instrumented fusion. Am J Infect Dis 2011;7:24–7. [18] Phillips MS, von Reyn CF. Nosocomial infections due to nontuberculous mycobacteria. Clin Infect Dis 2001;33:1363–74. [19] Astagneau P, Desplaces N, Vincent V, et al. Mycobacterium xenopi spinal infections after discovertebral surgery: investigation and screening of a large outbreak. Lancet 2001;358:747–51. [20] Grange JM. Complications of bacille Calmette-Guerin (BCG) vaccination and immunotherapy and their management. Commun Dis Public Health 1998;1:84–8. [21] Newport MJ, Huxley CM, Huston S, et al. A mutation in the interferon-gamma-receptor gene and susceptibility to mycobacterial infection. N Engl J Med 1996;335:1941–9. [22] Altare F, Durandy A, Lammas D, et al. Impairment of mycobacterial immunity in human interleukin-12 receptor deficiency. Science 1998;280:1432–5. [23] Newport MJ, Nejentsev S. Genetics of susceptibility to tuberculosis in humans. Monaldi Arch Chest Dis 2004;61:102–11. [24] Richter E, Weizenegger M, R€usch-Gerdes S, Niemann S. Evaluation of genotype MTBC assay for differentiation of clinical Mycobacterium tuberculosis complex isolates. J Clin Microbiol 2003;41: 2672–5. [25] Silber JS, Anderson DG, Vaccaro AR, et al, NASS. Management of postprocedural discitis. Spine J 2002;2:279–87.