Development of a combination antibiogram for empirical treatments of Pseudomonas aeruginosa at a university-affiliated teaching hospital.

Abstract

The significantly higher mortality rate in the critical illness patients with Pseudomonas aeruginosa (PA) infection is linked to inappropriate selecting of empirical treatment. Traditional local antibiogram provides clinicians the resistant rate of a single antimicrobial agent to the pathogen in the specific setting. The information is valuable to the clinicians in selecting suitable empirical antibiotic therapy. However, traditional local antibiogram can only provide information for single agent empirical antibiotic not combination regimens. The combination antibiogram should be developed to facilitate the selection of appropriate antibiotics to broader the coverage rate of resistant PA. The susceptibility to the β-lactam antibiotics (piperacillin/tazobactam (PTZ), ceftazidime, cefepime, imipenem, or meropenem) or to those administered in combination with an aminoglycoside (gentamicin or amikacin) or fluoroquinolone (ciprofloxacin or levofloxacin) was calculated. The chi-square test was used to compare the differences of combination coverage rates between non-ICU and ICU isolates. 880 PA isolates were isolated during study period. The susceptibility of single agents ranged from 83.1% to 89.7%. The combination regimens containing amikacin provide the highest cover rate (98.9%-99.1%) and those containing levofloxacin provide less coverage rate (92.3%-93.9%). The susceptibility to five β-lactam single agents in ICU isolates significantly lower than non-ICU isolates. The non-ICU isolates exhibited significantly higher susceptibility to the PTZ-gentamicin (p = 0.002) and ceftazidime-gentamicin (p = 0.025) than ICU isolates. Our results support the use of aminoglycosides instead of fluoroquinolones as additive agents in empirical combination treatments for patients with critical infections caused by PA.