<h3>BACKGROUND CONTEXT</h3> Vertebral osteomyelitis (VO) is a potentially devastating spinal infection associated with a significant morbidity burden and a mortality rate. The incidence of VO is increasing and its diagnosis is often delayed due to its characteristic non-specific, insidious onset of symptoms (recalcitrant back pain). Nonsurgical management consisting of a 6-week course of antibiotics with optional bracing remains first line treatment for VO; however, surgical intervention is often required. At present, data are limited on the epidemiology of surgical management in VO in the United States as previous reports are based on older studies with small sample sizes, mostly from Europe or Asia Pacific. Furthermore, no study to date has evaluated the impact of patient characteristics on surgical outcomes and complications on a large scale. <h3>PURPOSE</h3> To analyze the characteristics of patients who have VO requiring surgical intervention. <h3>STUDY DESIGN/SETTING</h3> National database study. <h3>PATIENT SAMPLE</h3> This study included 228,044 patients with vertebral osteomyelitis from 1998 to 2013 (44,401 surgically managed). <h3>OUTCOME MEASURES</h3> (1) Patient demographics: age, sex, race, Deyo Index (DI - a measure of comorbidity), type of comorbid pathology; (2) hospitalization data: length of stay (LOS), hospitalization charges, insurance provider; (3) surgical procedure type and utilization; (4) complication and mortality rate. <h3>METHODS</h3> A review of the National Inpatient Sample (NIS) database was performed for all patients admitted in the United States between 1998 to 2013 with a diagnosis of VO. Patients were then sub-stratified into those who underwent surgical intervention vs those treated with nonsurgical management. Patient demographics, hospitalization data, surgical procedure type and utilization, and complication and mortality rate were compared between groups. Differences between patient demographics for the surgical and nonsurgical groups were assessed using independent sample t-tests, linear regressions, and ANOVA. <h3>RESULTS</h3> A total of 228,044 patients were estimated to be admitted in the United States for VO from 1998 to 2013, of which 44,401 patients underwent surgical intervention. The incidence of surgical intervention for patients who had VO significantly increased from 0.6 per 100,000 U.S. persons in 1998 to 1.1 per 100,000 US persons in 2013. Compared to the nonsurgically managed cohort, surgically managed VO patients were significantly p <0.001) younger (56 surgical vs 60 non-surgical), had a lower mean DI (0.88 vs 1.33) and lower hospital mortality rate, but experienced a significantly (p <0.001) longer length of hospitalization (9.01 vs 9.98 days) and higher total hospitalization charges. Regardless of surgical approach, surgery of the thoracic spine carried the highest complication and mortality rate (49% and 3.4%, respectively). Anterior approach to the cervical spine was associated with the lowest rate of any complication (10.2%). Surgical management of VO of the lumbar spine had the lowest mortality rate for all approaches: anterior (0.3%), posterior (0.2%), or combined (0.1%). Congestive heart failure and liver disease most significantly predicted mortality (OR: 6.1 for CHF and OR: 2.9 liver disease) and renal disease most significantly predicted increased total hospital costs (OR: 7.4). The mean inflation adjusted total hospital costs increased from $20,355 per patient in 2001 to $39,991 in 2013. <h3>CONCLUSIONS</h3> VO is an understudied condition with a steady increase in its incidence in the US. It appears that the rate and the need for invasive procedures to salvage this condition is also increasing. This study identified certain patient demographics, comorbidities, and surgical approaches associated with higher complications, mortality rate and overall costs. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.
Read full abstract