Tuberculous vertebral osteomyelitis in the new millennium: still a diagnostic and therapeutic challenge

Abstract

In order to determine the clinical features and current prognosis of tuberculous vertebral osteomyelitis, the charts of all patients diagnosed with definite or probable tuberculous vertebral osteomyelitis from January 1983 to June 2002 ( n=78) were reviewed. The mean delay to diagnosis was 6.1 months. Sixty-five (83.3%) patients had inflammatory spinal pain, 35 (44.9%) had some neurological deficit, and only 27 (34.6%) had fever. Paravertebral, epidural, and psoas abscesses were detected in 73.1, 65.4, and 24.4% of the cases, respectively. Culture was positive in 48% of the percutaneous biopsies and in 61.7% of the open biopsies. After histological findings were included, the diagnostic yield of percutaneous biopsies was 68%. Fifty-five (70.5%) patients required surgical treatment at some stage of the disease. Although no deaths were directly attributable to tuberculous vertebral osteomyelitis and only 5.1% of patients relapsed, the mean overall hospital stay was 69.1+/-36.9 days, and 30 (38.5%) patients had severe functional sequelae. In conclusion, diagnosis of tuberculous vertebral osteomyelitis requires a high degree of suspicion. Percutaneous biopsy should be undertaken as soon as possible in any patient with compatible symptoms or radiological images in order to initiate suitable therapy.

Surgical treatment of spondylodiscitis. An update

“…Polymicrobial infections are uncommon and usually result from contiguous spread [54]. Tuberculosis is the most common cause of spondylodiscitis worldwide, even in developed countries where it represents up to 46% of all spondylodiscitis cases [9]. Fungal or parasitic infections are extremely uncommon [19].…”

Establishing the diagnosis of tuberculous vertebral osteomyelitis

“…It is possible that in the context of disseminated tuberculosis in immunosuppressed patients, many cases of TVO are not identified. In our experience, 19 % of patients with TVO were co-infected with HIV or had other types of immunosuppression [10]. Thus, in any immunosuppressed patient with back pain we should consider the possibility of TVO, regardless of whether the data are suggestive of disseminated tuberculosis.…”

“…In most studies, the time to diagnosis ranged between 4 and 6 months, with no difference between developed and developing countries [10,12,20,21].…”

“…By contrast, in developed countries, the age of the patient with spinal tuberculosis has a bimodal distribution with two peaks, one between 20 and 40 years of age and another between 60 and 80 years [10]. The first peak usually concerns immigrants or patients with HIV infection, and the second, immunosuppressed patients or patients with debilitating diseases.…”

“…Moreover, patients with a longer clinical course need surgical treatment more frequently than those with an earlier diagnosis [6,8]. Surprisingly, the incorporation of MRI into clinical practice has not significantly improved the diagnostic delay in TVO [9,10]. This suggests that physicians often remain unaware of this severe form of extrapulmonary tuberculosis.…”