196. Discontinuation of empirical vancomycin use in patients with suspected Methicillin-resistant Staphylococcus aureus (MRSA) pneumonia in intensive care units (ICUs)
Session: Poster Abstract Session: Antimicrobial Stewardship: Current State and Future Opportunities
Thursday, October 8, 2015
Room: Poster Hall
Background: Although the combination of respiratory tract and nasal cultures negative for MRSA has a high negative predictive value for MRSA infection, physicians are sometimes reluctant to discontinue empirical vancomycin, which is prescribed for suspected MRSA pneumonia in critically ill patients. We evaluated the clinical outcomes for discontinuing empirical vancomycin therapy in patients with suspected MRSA pneumonia at ICU admission and during ICU stay.

Methods: This study was conducted at four ICUs in two tertiary hospitals between Jan. 2010 and Dec. 2014. Among the adult patients who had been received empirical vancomycin for suspected MRSA pneumonia, the clinical characteristics and outcomes of patients in whom vancomycin was discontinued within 72 hours of documentation of negative for both respiratory tract and nasal cultures for MRSA (VAN-DC group) were compared with those in whom vancomycin was continued (VAN group). 

Results: Of 93 patients with respiratory tract and nasal cultures negative for MRSA, 19 (20%) were VAN-DC group and 74 (80%) were VAN group. Among the 93 patients, empirical vancomycin had been prescribed to 26 (28%) at ICU admission, and to 67 (72%) at a median of 8 days after ICU admission. The median duration of vancomycin prescription was 5 days in VAN-DC group and 14 days in VAN group. Carbapenem (78%) is the most commonly co-administered antibiotics with vancomycin. The 14 day- and 30 day-mortality in VAN-DC group was similar to those in VAN group after diagnosis of pneumonia (21.1% vs. 17.6%, P = 0.73; 31.6% vs. 40.5%, P = 0.47). There were no significant differences in the median APACHE II score on the day of vancomycin initiation (19 vs. 16, P = 0.09), rate of acute kidney injury (35.7% vs. 45.1%, P = 0.69), positive results for other pathogens (52.7% vs. 52.6%, P = 0.99), and the median ICU length of stay (28 days vs. 22 days, P = 0.44) between VAN-DC and VAN group. However, a modified clinical pulmonary infection score (CPIS) of ≤ 6 on day 3 (94.7% vs. 71.6%, P = 0.04) and formal ID consultation (56.3% vs. 25.6%, P = 0.03) were associated with VAN-DC group.

Conclusion: Continued empirical vancomycin use with negative cultures for MRSA was a common practice in ICUs, which was not helpful to decrease the mortality of the patients with suspected MRSA pneumonia.

Oh-Hyun Cho, MD1, Ki-Ho Park, MD2, Wonyong Jo, MD3, Mi Suk Lee, MD, PhD2 and In-Gyu Bae, MD4, (1)Department of Internal Medicine, Gyeongsang National University Hospital, Jinju, South Korea, (2)Division of Infectious Diseases, Department of Internal Medicine, Kyung Hee University Hospital, Seoul, South Korea, (3)Department of Internal Medicine, Gyeongsang National Univesity Hospital, Jinju, South Korea, (4)Division of Infectious Diseases Department of Internal Medicine, Gyeongsang National University Hospital, JinJu, South Korea


O. H. Cho, None

K. H. Park, None

W. Jo, None

M. S. Lee, None

I. G. Bae, None

Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 7th with the exception of research findings presented at the IDWeek press conferences.