MULTIPLY ANTIBIOTIC-RESISTANT GRAM-NEGATIVE BACILLI IN A LONG-TERM-CARE FACILITY - A CASE-CONTROL STUDY OF PATIENT RISK-FACTORS AND PRIOR ANTIBIOTIC USE
Rr. Muder et al., MULTIPLY ANTIBIOTIC-RESISTANT GRAM-NEGATIVE BACILLI IN A LONG-TERM-CARE FACILITY - A CASE-CONTROL STUDY OF PATIENT RISK-FACTORS AND PRIOR ANTIBIOTIC USE, Infection control and hospital epidemiology, 18(12), 1997, pp. 809-813
OBJECTIVE: To determine the relation between prior exposure to specifi
c antimicrobials and acquisition of gram-negative bacilli resistant to
multiple beta-lactam and aminoglycoside antibiotics among long-term-c
are patients. DESIGN: Case-control study. Cases were patients from who
m multiply resistant Enterobacteriaceae or Pseudomonas aeruginosa were
isolated; controls were patients from whom nonresistant bacteria of t
he same species were isolated. Prospectively defined risk factors incl
uded underlying illness, activity level, presence of decubitus ulcers,
presence of indwelling devices, and prior exposure to specific antimi
crobial agents. Resistant and control isolates of P aeruginosa were co
mpared using pulsed-field gel electrophoresis (PFGE) of genomic DNA af
ter digestion with XbaI. SETTING: 390-bed long-term Veterans' Affairs
facility. RESULTS: We identified 35 patients with multiply resistant E
nterobacteriaceae and 24 patients with multiply resistant P aeruginosa
. Of the resistant Enterobacteriaceae, 87% of isolates were resistant
to piperacillin, 55% to ceftazidime, and 90% to gentamicin. Acquisitio
n of multiply resistant Enterobacteriaceae was associated with presenc
e of decubitus ulcers (odds ratio [OR], 12.2; 95% confidence interval
[CI95], 3.3-44.2; P=.0002) and prior receipt of ampicillin (OR 13.7; C
I95, 2.2-84; P=.005). Of resistant isolates of P aeruginosa, 88% were
resistant to piperacillin, 25% to ceftazidime, 42% to imipenem, and 67
% to ciprofloxacin. Isolation of a multiply resistant P aeruginosa was
associated with total days of antimicrobial exposure (OR 1.07; CI95,
1.01-1.12; P=.011) and not with prior receipt of any individual agent.
Eleven multiply resistant isolates shared a common PFGE pattern. CONC
LUSIONS: In our long-term-care facility, acquisition of multiply resis
tant Enterobacteriaceae was associated with the presence of decubitus
ulcers and prior exposure to ampicillin. Acquisition of resistant P ae
ruginosa was associated with total antibiotic exposure. Molecular typi
ng of P aeruginosa isolates implicated patient-to-patient transmission
of a limited number of resistant strains (Infect Control Hosp Epidemi
ol 1997;18:809-813).