Background: The national dialog surrounding the opioid epidemic has focused on the mortality of overdose while other medical consequences of this epidemic remain poorly characterized. This 11 year retrospective chart review seeks to better describe the epidemiology and clinical features of native valve infectious endocarditis (IE) stratified by drug use.
Methods: IE cases were identified using ICD-9 codes from patients discharged from WFUBMC between January 2004-September 2015. Cases with cardiac devices, previous endocarditis, and prosthetic valves were excluded. Self-reported drug use with drug name and route of abuse explicitly stated were required for classification as injection drug users (IDU). Data were analyzed with t tests, chi-square tests, and Fishers exact tests.
Results: 277 IE cases were identified: 188 in non-injectors (non-IDU) and 89 in IDU with an increasing proportion of cases attributable to IDU over time. Oxymorphone became the most frequently reported drug of abuse (Fig. 1). The IDU group was younger than the non-IDU group (average age of 31.8 ± 9.3 vs 54.2 ± 15.4 respectively, p<0.001), more likely to be Caucasian (95.5% vs 76.1%, p<0.001) and to be female (51.7% vs 36.2%, p<0.05). The IDU group was less likely to have mitral (20.2% IDU vs 43.6% non-IDU, p<0.01) or aortic valve disease (12.4% IDU vs 34.6% non-IDU, p<0.01), have cerebral emboli (7.9% vs 27.1%, p<0.001) and to die (5.6% vs 23.4%, p<0.001). There was no difference in length of stay (20.8 ± 14.8 IDU vs. 17.6 ± 13.2 non-IDU, p=0.09), need for intensive care (61.8% IDU vs 66.8% non-IDU, p=0.41) or dialysis initiation (11.2% IDU vs 14.9% non-IDU, p=0.57). While the non-IDU group trended towards more frequent valve replacement (p=0.055), there was no difference in the number of cardiac surgeries when valve replacement and repair were taken together (23.6% for IDU vs 29.3% for non-IDU, p0.32). Staphylococcus aureus was the most common cause in both groups (76.4% IDU and 40.5% non-IDU).
Conclusion: WFUBMC is experiencing a rise in the proportion of IE cases associated with IDU. The demographics of IDU-associated cases differ substantially from non-IDU cases but clinical morbidity of IDU-associated IE is surprisingly severe and equivalent to non-IDU IE by many clinical parameters.
L. Bachmann, Cepheid: Research Contractor , Research support
See more of: Poster Abstract Session