Pathogen Identification in Suspected Cases of Pyogenic Spondylodiscitis Publisher Pubmed



Sheikh AF^{1, 2} ; Khosravi AD^{1, 2} ; Goodarzi H^{1, 2, 3} ; Nashibi R^{2, 4} ; Teimouri A^{5, 6} ; Motamedfar A⁷ ; Ranjbar R³ ; Afzalzadeh S⁴ ; Cyrus M² ; Hashemzadeh M^{1, 2} Show Affiliations
Source: Frontiers in Cellular and Infection Microbiology Published:2017

Abstract

Pyogenic spinal infection continues to represent a worldwide problem. In approximately one-third of patients with pyogenic spondylodiscitis, the infectious agent is never identified. Of the cases that lead to organismal identification, bacteria are more commonly isolated from the spine rather than fungi and parasites. This study applied universal prokaryotic 16S rRNA PCR as a rapid diagnostic tool for the detection of bacterial agents in specimens from patients suspected of pyogenic spondylodiscitis. Gram and Ziehl-Neelsen staining were used as a preliminary screening measure for microbiologic evaluation of patient samples. PCR amplification targeting 16S rRNA gene was performed on DNA extracted from 57 cases including specimens from epidural abscesses, vertebral, and disc biopsies. Positive samples were directly sequenced. MRI findings demonstrated that disc destruction and inflammation were the major imaging features of suspected pyogenic spondylodiscitis cases, as 44 cases showed such features. The most common site of infection was the lumbar spine (66.7%), followed by thoracic spine (19%), the sacroiliac joint (9.5%), and lumbar-thoracic spine (4.8%) regions. A total of 21 samples amplified the 16S rRNA-PCR product. Sanger sequencing of the PCR products identified the following bacteriological agents: Mycobacterium tuberculosis (n = 9; 42.9%), Staphylococcus aureus (n = 6; 28.5%), Mycobacterium abscessus (n = 5; 23.8%), and Mycobacterium chelonae (n = 1; 4.8%). 36 samples displayed no visible 16S rRNA PCR signal, which suggested that non-bacterial infectious agents (e.g., fungi) or non-infectious processes (e.g., inflammatory, or neoplastic) may be responsible for some of these cases. The L3–L4 site (23.8%) was the most frequent site of infection. Single disc/vertebral infection were observed in 9 patients (42.85%), while 12 patients (57.15%) had 2 infected adjacent vertebrae. Elevated erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) inflammatory markers were noted in majority of the patients. In conclusion, microbiological methods and MRI findings are vital components for the proper diagnosis of pyogenic spondylodiscitis. Our findings suggest that molecular methods such as clinical application of 16S rRNA PCR and sequencing may be useful as adjunctive diagnostic tools for pyogenic spondylodiscitis. The rapid turnaround time of 16S rRNA PCR and sequencing submission and results can potentially decrease the time to diagnosis and improve the therapeutic management and outcome of these infections. Although S. aureus and M. tuberculosis were the most common causes of pyogenic spinal infections in this study, other infectious agents and non-infectious etiologies should be considered. Based on study results, we advise that antibiotic therapy should be initiated after a definitive etiological diagnosis. © 2017 Sheikh, Khosravi, Goodarzi, Nashibi, Teimouri, Motamedfar, Ranjbar, Afzalzadeh, Cyrus and Hashemzadeh.