Juan D. Colmenero, Juan D. Ruiz-Mesa, Rocío Sanjuan-Jimenez, Beatriz Sobrino, Pilar Morata


June 2013, Volume 22, Issue 4, pp 579 - 586 Review Article Read Full Article 10.1007/s00586-012-2348-2

First Online: 11 May 2012

Purpose

The aim of this article has been to analyze the clinical and radiological data suggesting tuberculous vertebral osteomielitis (TVO), and then discuss the steps to be followed to achieve an aetiological diagnosis.

Methods

A thorough literature search was carried out to identify the best clinical and microbiological evidence for a fast and efficient diagnosis of TVO.

Results

The clinical and radiological diagnosis of spinal tuberculosis suffers from serious limitations, with a high percentage of cases requiring vertebral biopsy to reach a definitive diagnosis. The increasing incidence of multidrug-resistant tuberculosis has highlighted the insufficiency of the histopathological diagnosis and the need for microbiological diagnosis. Unfortunately, the maximum sensitivity of spinal tuberculosis cultures is 80 %, and traditional methods require 6 to 8 weeks for the isolation, identification and sensitivity study. New culture media and identification methods have improved sensitivity and reduced the time required for the identification. Molecular methods have now been integrated into a single test, with identification of the mycobacterium responsible and its sensitivity to rifampicin. Additionally, multiplex-PCR tests have been developed that allow a rapid differential diagnosis between granulomatous spondylodiscitis.

Conclusions

All patients with subacute inflammatory back or neck pain showing suggestive radiological findings should be studied to rule out TVO. If there is no clear evidence of tuberculosis from another location or indication for surgery, a percutaneous vertebral biopsy should be performed. When TVO is suspected, all spinal or paravertebral tissue samples should be sent simultaneously to pathology and microbiology laboratories for appropriate processing.


Read Full Article