855Risk Factors for Peripherally Inserted Central Catheter (PICC) Bloodstream Infection (BSI)

Background. PICC use may have lower risk of BSI compared to other central catheters but little data exist regarding risks for PICC associated BSI. This study evaluates risks for PICC BSI among patients (pts) at an academic hospital. Methods. PICC BSI rate was calculated among all pts with PICCs inserted between July 2009-July 2012. A nested case control study (1:3 ratio of PICC BSI cases to no BSI controls) matched for operator (interventional radiology [IR], Infectious Diseases [ID], or nurse venous access team [VAT]) was conducted to assess risks for PICC BSI. Pt characteristics, PICC type (antimicrobial impregnated [ABXI] or not) and indication for PICC were recorded from chart review. Logistic regression was used. Results. 89 PICC BSI occurred among 5372 PICC placements over the study (rate 1.66%). Mean days to infection were 32 and this was not different dependent on use of an ABXI PICC. Having an IR operator was associated with a 1.75 fold higher risk for PICC BSI compared to ID (p = 0.02) and a 4.22 fold higher risk compared to VAT (p = 0.0008). There was no difference in risk comparing ID to VAT operators. After multivariate analysis, non-ABXI PICC, PICC inserted for chemotherapy (CXT), tunneled PICC, leukemia, and AIDS were significant independent risks for PICC BSI (Table). Logistic Regression Analysis of Risk Factors for PICC BSI Conclusion. PICC BSI rate was lowest if the PICC operator was ID or VAT. Use of a non ABXI PICC significantly increased risk for PICC BSI. Highest risk was among pts with AIDS and leukemia and if the indication for PICC was CXT. Tunneled PICCs were placed for pts at high risk for PICC infection or complications. Given our findings, catheter tunneling should be studied further to determine if PICC BSI risk is mitigated. Disclosures. All authors: No reported disclosures.