A 9-Year retrospective review of antibiotic cycling in a surgical intensive care unit

Abstract

Background: Six years after initiating a monthly antibiotic cycling protocol in the surgical intensive care unit (SICU), we retrospectively reviewed antibiogram-derived sensitivities of predominant gram-negative pathogens before and after antibiotic cycling. We also examined susceptibility patterns in the medical intensive care unit (MICU) where antibiotic cycling is not practiced.

Materials and methods: Antibiotic cycling protocol was implemented in the SICU starting in 2003, with monthly rotation of piperacillin/tazobactam, imipenem/cilastin, and ceftazidime. SICU antibiogram data from positive clinical cultures for years 2000 and 2002 were included in the pre-cycling period, and those from 2004 to 2009 in the cycling period.

Results: Profiles of SICU pseudomonal isolates before (n = 116) and after (n = 205) implementing antibiotic cycling showed statistically significant improvements in susceptibility to ceftazidime (66% versus 81%; P = 0.003) and piperacillin/tazobactam (75% versus 85%; P = 0.021), while susceptibility to imipenem remained unaltered (70% in each case; P = 0.989). Susceptibility of E. coli isolates to piperacillin/tazobactam improved significantly (46% versus 83%; P < 0.0005), trend analysis showing this improvement to persist over the study period (P = 0.025). Similar findings were not observed in the MICU. Review of 2004-2009 antibiotic prescription practices showed monthly heterogeneity in the SICU, and a 2-fold higher prescribing of piperacillin/tazobactam in the MICU (P < 0.0001).

Conclusions: Six years into antibiotic cycling, we found either steady or improved susceptibilities of clinically relevant gram-negative organisms in the SICU. How much of this effect is from cycling is unknown, but the antibiotic heterogeneity provided by this practice justifies its ongoing use.

Figure 1

Trend in antimicrobial susceptibilities of common gram-negative isolates: (A) Pseudomonas aeruginosa; (B) Escherichia coli; and (C) K. pneumoniae. CAZ = ceftazidime; IMI = imipenem/cilastin; PIP/T = piperacillin/tazobactam

Fig. 2

Average monthly prescription frequency of each of the three target drugs from 2004 to 2009 when designated as AOTM (4 mo yearly per study antibiotic), and when not the designated AOTM. Compared with periods as AOTM, the overall prescription frequency of target antibiotics fell by 63% when not AOTM. AOTM =antibiotic of the month.

Fig. 3

Antibiotic prescription patterns in SICU (top) and MICU (bottom) from 2004 to 2009.