Treatment Failure in Vertebral Osteomyelitis: Is it All About Staphylococcus aureus ?

Retrospective cohort study.

The aim was to compare the influence of 2 common vertebral osteomyelitis (VO) causing pathogens on treatment failure within the first year of diagnosis.

VO is mainly caused by Staphylococcus aureus (SA), while enterococci and streptococci (ENST) are also responsible for a significant proportion of VO, particularly in elderly patients. Data on VO caused by SA show a tendency for worse outcome, whereas data on VO caused by ENST are scarce. For this purpose, our study compares characteristics of patients with VO caused by SA or ENST in order to analyze risk factors for treatment failure.

We conducted a retrospective monocentric study including VO patients from 2008 to 2020. Primary outcome was treatment failure defined as death or relapse within 1 year (T1). We compared patients diagnosed with VO caused by Staphylococcus aureus including MRSA to patients diagnosed with VO caused by Enterococcus and Streptococcus species, which were combined into one group. Polymicrobial infections were excluded. We employed multiple logistic regression analysis to adjust for confounding. To account for moderation, the model was repeated with an included interaction term.

Data of 130 VO patients (SA=95; ENST=35) were available at T1. Treatment failure occurred in 37% of SA patients and 23% of ENST patients. On multivariate analysis SA [odds ratio (OR): 3.12; 95% confidence interval (CI): 1.09-10.53; P =0.046], Charlson comorbidity index (OR: 1.31; 95% CI: 1.11-1.58; P =0.002) and infectious endocarditis (IE; OR: 4.29; 95% CI: 1.23-15.96; P =0.024) were identified as independent risk factors for treatment failure.

In our cohort every third patient with VO caused by SA or ENST dies within 1 year. Our findings indicate that patients with VO caused by SA, concomitant IE and/or a high Charlson comorbidity index score may be at elevated risk for treatment failure. These findings can be used to individualize patient care and to direct clinical surveillance. This could include echocardiography evaluating for the presence of IE in patients with VO caused by gram-positive pathogens.