(Chest. 2001;119:426S-430S.)
© 2001
American College of Chest Physicians
Antibiotic Utilization*
Is There an Effect on Antimicrobial Resistance?
Jan E. Patterson, MD
*
From the Department of Medicine, University of Texas Health Science Center, San Antonio, TX.
Correspondence to: Jan E. Patterson, MD, University of Texas Health Science Center, Department of Medicine, 7703 Floyd Curl Dr, Mail Code 7881, San Antonio, TX 78229-3900; e-mail: pattersonj{at}uthscsa.edu
 |
Abstract
|
|---|
The antimicrobial resistance problem in hospitals continues to
worsen. In particular, extended-spectrum ß-lactamase–producing
Klebsiella pneumoniae (ESBL-KP) and vancomycin-resistant
enterococci (VRE) are significant causes of morbidity and mortality
among critically ill patients. Treating infections caused by these
pathogens presents therapeutic dilemmas. The association between
broad-spectrum ß-lactam overutilization and selection for ESBL-KP has
been appreciated for some time; several institutions have reported a
decrease in the prevalence of ESBL-KP with a shift in antibiotic
utilization from third-generation cephalosporins to other
broad-spectrum drugs. Currently, optimal treatment of ESBL-KP includes
the carbapenems, but widespread use of these drugs is expensive and may
be associated with further selection of antibiotic resistance and/or
superinfection with other inherently resistant pathogens. VRE are
especially difficult organisms to treat because of their inherent and
acquired resistance to most currently available antibiotics. The
prevalence of VRE has also been documented to decrease upon a shift in
antibiotic use from third-generation cephalosporins to broad-spectrum
antibiotics of other classes. Thus, antibiotic utilization measures
appear to contribute to the control of the emergence of
multidrug-resistant pathogens such as ESBL-KP and
VRE.
Key Words: antibiotic • antimicrobial • extended-spectrum ß-lactamase • multidrug resistance • outbreak • pathogen • resistance • vancomycin
|
Introduction
|
|---|
In
1992, nosocomial infections were reported to cause > 8 million excess
hospital days, approximately 80,000 deaths, and a burden of $4.5
billion on the health-care system.1
The factors
responsible for increasing antibiotic resistance are listed in Table 1
. Antibiotic-resistant pathogens such as extended-spectrum
ß-lactamase-producing Klebsiella pneumoniae (ESBL-KP) and
vancomycin-resistant enterococci (VRE) are important causes of
nosocomial infections among critically ill, hospitalized patients.
Outbreaks of ESBL-KP often involve the emergence of several different
strains that have acquired plasmids encoding multidrug
resistance.2
VRE are commonly placed into two types of
resistance classes, termed vanA or vanB. Resistance of the vanA
phenotype is acquired from a transferable plasmid, whereas vanB
resistance may be chromosomally mediated or transferable.
Emergence of the vanA phenotype may be clonal but is often polyclonal,
whereas emergence of the vanB phenotype may be polyclonal but is often
clonal.3
The presence of genes encoding multidrug
resistance limits therapeutic options in many cases.
A strong correlation between the use of third-generation
cephalosporins, such as ceftazidime, and antibiotic resistance in
K pneumoniae has been repeatedly
demonstrated.2
4
5
6
7
In addition, a number of
studies3
8
9
10
have also shown an association
between broad-spectrum cephalosporin use and VRE. While antibiotic use
in general cannot always be correlated with the emergence of antibiotic
resistance, the association of cephalosporin use with these two problem
pathogens has been the focus of several
studies.2
3
4
5
6
7
8
9
10
11
A decline in the prevalence of
ESBL-KP and VRE has been documented upon implementing an
institution-wide shift to other classes of broad-spectrum
antipseudomonal agents. The purpose of this article is to review
existing data implicating antibiotic use in increasing the prevalence
of ESBL-KP and VRE. In addition, evidence that decreasing
third-generation cephalosporin use has been associated with significant
decreases in the rates of ESBL-KP and VRE isolation will be presented.
|
Is There a Correlation Between Antibiotic Resistance and Mortality?
|
|---|
The presence of antibiotic resistance in a bacterial strain is
clearly a factor in increasing mortality rates. At New York Hospital
Medical Center (Queens, NY), 23 of 43 evaluable patients with
ceftazidime-resistant K pneumoniae (CRKP)
died.4
Paterson and colleagues12
demonstrated
the association between ESBL-KP and mortality in an ongoing prospective
multicenter study taking place in seven hospitals on six continents. Of
216 cases of K pneumoniae bacteremia, 15% were caused by
ESBL-KP. Patients with ESBL-KP who received empiric treatment to which
their strains were resistant had a significantly higher rate of
mortality (75%) compared with patients who received appropriate
antibiotics (28%; p = 0.02). These data illustrate the importance of
a correct empiric antibiotic choice for critically ill patients.
At Johns Hopkins Hospital, patients with VRE had significantly
increased lengths of stay in the hospital and ICU and higher crude
mortality rates (45% vs 27%; p = 0.007) compared with patients with
vancomycin-susceptible enterococci (VSE).13
Similarly, at
the University of Pittsburgh Medical Center, Linden et
al14
found an association between VRE and mortality among
liver transplant patients. In this study, 46% of patients with VRE
died from enterococcal bacteremia compared with 25% of patients with
VSE (p = 0.04). Confirming observations at Johns Hopkins, Linden et
al14
found that patients with VRE had significantly longer
hospital stays than patients with VSE (p = 0.03).
|
Is Antibiotic Use a Risk Factor for Increased ESBL-KP and VRE
Prevalence?
|
|---|
In a presentation by Paterson and coworkers,12
31%
of patients with ESBL-KP had received a third-generation
cephalosporin at least 14 days preceding bacteremia compared with 3%
of non-ESBL-KP patients (p < 0.01). Rice and
colleagues5
showed that ceftazidime use was a risk factor
for increased prevalence of CRKP in the Cleveland Department of
Veterans Affairs Medical Center between January and December 1994.
Notably, there was a strong association between the amount (in grams)
of ceftazidime used in a particular ward and the prevalence of CRKP in
that ward (p = 0.002). At New York Hospital Medical Center, CRKP
comprised 17.3% of all K pneumoniae isolates recovered
between October 1988 and April 1990. Interestingly, 91 of 127 patients
(72%) with CRKP had received > 7 days of antibiotic treatment; 41%
of CRKP isolates were from patients previously treated with
ceftazidime.4
Furthermore, there was a strong correlation
between prior ceftazidime use and serious infections caused by
CRKP.4
A study by Tokars and coworkers8
from the Atlanta
Veterans Affairs Medical Center showed that exposure to
broad-spectrum antimicrobials that lack enterococcal activity,
including cephalosporins (ceftriaxone, ceftazidime, vancomycin,
clindamycin, cefuroxime, and ticarcillin-clavulanic acid), was strongly
associated with VRE colonization. In addition, the number of days in
which patients received antimicrobials correlated with VRE prevalence.
In 1997, Gerding15
reviewed the literature regarding the
relationship between prior antibiotic use and the emergence of VRE. Of
18 studies, 15 studies implicated vancomycin use in the emergence of
VRE, 4 of 6 studies implicated cephalosporins, 4 of 5 studies
implicated metronidazole, and 5 of 6 studies implicated use of any type
of antimicrobial.15
A case-control study3
of
seven hospitals in the San Antonio, TX, area determined that the risk
factors for acquiring VRE as compared with VSE were prior use of
third-generation cephalosporins (p = 0.03), parenteral vancomycin
(p = 0.002), or any antibiotics (p = 0.04).
A case-control study by Pallares and colleagues16
sought
to determine whether the use of second-generation and third-generation
cephalosporins is a risk factor for nosocomial bacteremia due to
Enterococcus faecalis. Included in this study16
were 207 patients with nosocomial enterococcal bacteremia, of which 156
patients (75.4%) were matched with nonbacteremic control patients
based on age, sex, date of admission, hospital service, days of
hospitalization, primary diagnoses, and similar operative procedures.
Patients who had contracted nosocomial enterococcal bacteremia were
more likely to have been treated with second-generation and
third-generation cephalosporins (odds ratio, 4.8; 95% confidence
interval, 2.3 to 9.8),16
while prior use of other
antibiotics did not appear to increase the risk of contracting
nosocomial enterococcal bacteremia.16
These data indicate
that cephalosporins may allow enterococcal overgrowth.16
Dahms and coworkers9
reported that use of either
third-generation cephalosporins or vancomycin is a risk factor for
postoperative infections due to VRE. In this study,9
surgical patients with VRE infections were matched with similar
patients with VSE infections. Patients with VRE received
third-generation cephalosporins, vancomycin, and a combination of both
for significantly more days than patients with VSE. The risk for VRE
infection was 1.6 for patients with prior use of third-generation
cephalosporins, 1.8 for vancomycin, and 2.7 for concurrent use of both.
Since existing evidence points to a role of cephalosporin use in the
increasing prevalence of ESBL-KP and VRE, mandating the decreased use
of cephalosporins may help to control outbreaks of these
antibiotic-resistant organisms.
Finally, Donskey and colleagues10
recently published data
from examiners in northeast Ohio who experienced an outbreak of VRE.
Five teaching hospitals were surveyed, and it was found that the rates
of isolation of VRE correlated with ticarcillin/clavulanate use
(p = 0.005). Third-generation cephalosporins and clindamycin had a
positive, but not significant, correlation.
|
Can Limiting Third-Generation Cephalosporin Use Reduce the
Prevalence of ESBL-KP and VRE?
|
|---|
A study performed by Rice and colleagues5
demonstrated that a decrease in ceftazidime utilization and concurrent
increase in piperacillin/tazobactam use correlated with a decrease in
CRKP without a concurrent increase in
piperacillin/tazobactam-resistant strains. Furthermore, their data
indicated that ceftazidime use increased resistance to other agents,
while increased use of piperacillin/tazobactam did not show evidence of
selecting for resistance either to other agents or to
piperacillin/tazobactam.5
A study at New York Hospital
also found a decrease in the rate of ESBL-KP upon a shift in antibiotic
utilization from cephalosporins to other antibiotics.6
An 80.1% reduction in hospital-wide cephalosporin use correlated
with a 44.0% reduction in CRKP infection and colonization throughout
the hospital (p < 0.01), a 70.9% reduction in all ICUs
(p < 0.001), and an 87.5% reduction in the surgical ICU
(p < 0.001).6
However, increased imipenem/cilastatin
use in this study was associated with an increased incidence of
imipenem-resistant Pseudomonas aeruginosa.6
A significant decrease in multidrug-resistant K
pneumoniae at our two university-affiliated hospitals in Texas was
observed after a decrease in ceftazidime use and an increase in
piperacillin/tazobactam use.7
Piperacillin/tazobactam
resistance decreased despite the increase in piperacillin/tazobactam
use. In addition to monitoring the decrease in multidrug-resistant
K pneumoniae, we have also monitored the prevalence of
infections due to Gram-negative bacilli that are resistant to
third-generation cephalosporins. Decreased cephalosporin use and
increased piperacillin/tazobactam use has been associated with a
decrease in infections caused by cephalosporin-resistant Gram-negative
bacilli in our adult ICUs (Fig 1
, top and bottom).
View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Increased piperacillin/tazobactam use with a
concurrent decrease in cephalosporin use is associated with a decrease
in infections due to cephalosporin-resistant Gram-negative bacilli in
adult ICUs. Top: Third-quarter 1999 antibiotic usage
rates (antibiotic days per 1,000 patient-days) in adult ICUs. The
decrease in cephalosporin use and increase in piperacillin/tazobactam
use over an extended time is also documented.7
Bottom: Prevalence of infections with
cephalosporin-resistant Gram-negative bacilli (infection rate per 1,000
patient-days) in the medical/coronary ICU (MI/CICU), surgical ICU
(SICU), and transplant ICU (TICU).
|
|
Infection control practices to combat VRE have been implemented with
varying success at several institutions. Two studies published by Boyce
and colleagues17
18
documented that barrier precautions
were effective at controlling two consecutive clonal strain outbreaks
of vanA and vanB class vancomycin-resistant Enterococcus
faecium. However, at the University of Maryland, when barrier
precautions and vancomycin restriction were instituted to control
a VRE outbreak, the prevalence of polyclonal VRE did not decrease
between 1993 (16.9%) and 1994 (18.7%).19
Because of the
emergence of polyclonal VRE in many institutions, other measures
besides infection control, such as antibiotic utilization, may be
needed to control antibiotic selection pressure for VRE.
At the Brooklyn Veterans Affairs Medical Center in New York, limiting
hospital-wide use of cefotaxime, ceftazidime, vancomycin, and
clindamycin was associated with a significant decrease in the
prevalence of fecal colonization with VRE from 47 to 15%
(p < 0.001).11
A study by Bradley et al20
demonstrated that restriction of third-generation cephalosporins in
febrile, neutropenic patients was associated with a reduction in VRE
prevalence. In phase 1 of this study, there was no change in antibiotic
administration, with ceftazidime prescribed as empiric therapy for
febrile, neutropenic patients. Phases 2a and 2b consisted of two 2- to
4-month intervals in which piperacillin/tazobactam was substituted for
ceftazidime. Finally, in phase 3, ceftazidime was substituted for
piperacillin/tazobactam for 4 months. The results of this study are
summarized in Table 2 . Piperacillin/tazobactam use significantly decreased the prevalence of
VRE colonization, and this decrease was promptly reversed upon return
to ceftazidime utilization.
In our bone marrow transplant unit at the Department of Veterans
Affairs hospital, we have also seen a decrease in VRE
colonization and infection after reemphasizing infection control
measures and changing our empiric regimen for febrile neutropenic
patients from ceftazidime to
piperacillin/ tazobactam.21
|
Conclusion
|
|---|
Antibiotic usage patterns exert a significant influence over the
rates of resistance observed in problematic multidrug-resistant
nosocomial pathogens. Emergence and spread of ESBL-KP are clearly
promoted by widespread use of extended-spectrum cephalosporins,
especially ceftazidime. Emergence of VRE within institutions is
promoted both by poor infection control techniques and by use of
broad-spectrum antimicrobial agents. In particular, increased rates of
VRE colonization and infection have been associated with use of
vancomycin, extended-spectrum cephalosporins, and antianaerobic drugs.
Strict adherence to well-accepted infection control guidelines, along
with caution in use of broad-spectrum antimicrobial agents, represents
the best strategy for preventing the emergence and spread of nosocomial
multidrug-resistant pathogens.
|
Appendix 1
|
|---|
Dr. Joseph Lynch:
Your data, like those of other
investigators, show that shifting away from ceftazidime to
piperacillin/tazobactam really makes a difference and that reduction in
antibiotic resistance is sustained. Have others seen a sustained
reduction?
Dr. Jan Patterson:
Dr. Louis Rice continued to see the
decline, and our resistance rates have continued to drop. Klebsiella
seems to be most amenable to the switch, but we are seeing declines in
some other organism resistance rates as well. We are also seeing less
VRE since the change.
Dr. George Eliopoulos:
Why do you think that your rates of
multidrug-resistant K pneumoniae declined even though
you had an organism that was not susceptible to
piperacillin/tazobactam?
Dr. Patterson:
One of our hospitals had primarily ESBL-KP, and
the other hospital had primarily K pneumoniae with a
type-1 ß-lactamase enzyme. Decreasing ceftazidime use, along with the
infection control measures, was associated with a decline in
ceftazidime and piperacillin/tazobactam resistance.
Dr. Rekha Murthy:
What are you giving neutropenic,
nontransplant patients?
Dr. Patterson:
Piperacillin/tazobactam plus or minus
gentamicin or cefipime. In our Veterans Affairs hospital,
piperacillin/tazobactam has slightly better activity against
Pseudomonas than ceftazidime and is better than imipenem. We have
recently started cycling cefipime and piperacillin/tazobactam in this
population.
|
Footnotes
|
|---|
Abbreviations:
CRKP = ceftazidime-resistant Klebsiella pneumoniae;
ESBL-KP = extended-spectrum ß-lactamase-producing Klebsiella
pneumoniae; VRE = vancomycin-resistant enterococci;
VSE = vancomycin-susceptible enterococci
|
References
|
|---|
-
Martone WJ, Jarvis WR, Culver DH, et al. Incidence and nature of endemic and epidemic nosocomial infections. In: Bennett JV, Brachman PS, eds. Hospital infections. 3rd ed. Boston, MA: Little, Brown and Company, 1992; 577–596
-
Rice, LB, Willey, SH, Papanicolaou, GA, et al (1990) Outbreak of ceftazidime resistance caused by extended-spectrum ß-lactamases at a Massachusetts chronic-care facility. Antimicrob Agents Chemother 34,2193-2199[Abstract/Free Full Text]
-
Moreno, F, Grota, P, Crisp, C, et al (1995) Clinical and molecular epidemiology of vancomycin-resistant Enterococcus faecium during its emergence in a city in southern Texas. Clin Infect Dis 21,1234-1237[ISI][Medline]
-
Meyer, KS, Urban, C, Eagan, JA, et al (1993) Nosocomial outbreak of Klebsiella infection resistant to late-generation cephalosporins. Ann Intern Med 119,353-358[Abstract/Free Full Text]
-
Rice, LB, Eckstein, EC, DeVente, J, et al (1996) Ceftazidime-resistant Klebsiella pneumoniae isolates recovered at the Cleveland Department of Veterans Affairs Medical Center. Clin Infect Dis 23,118-124[ISI][Medline]
-
Rahal, JJ, Urban, C, Horn, D, et al (1998) Class restriction of cephalosporin use to control total cephalosporin resistance in nosocomial Klebsiella. JAMA 280,1233-1237[Abstract/Free Full Text]
-
Patterson, JE, Hardin, TC, Kelly, CA, et al (2000) Association of antibiotic utilization measures and control of multiple-drug resistance in Klebsiella pneumoniae. Infect Control Hosp Epidemiol 21,455-458[CrossRef][ISI][Medline]
-
Tokars, JI, Satake, S, Rimland, D, et al (1999) The prevalence of colonization with vancomycin-resistant Enterococcus at a Veterans’ Affairs institution. Infect Control Hosp Epidemiol 20,171-175[CrossRef][ISI][Medline]
-
Dahms, RA, Johnson, EM, Statz, CL, et al (1998) Third-generation cephalosporins and vancomycin as risk factors for postoperative vancomycin-resistant Enteroccocus infection. Arch Surg 133,1343-1346[Abstract/Free Full Text]
-
Donskey, CJ, Schreiber, JR, Jacobs, MR, et al (1999) A polyclonal outbreak of predominantly VanB vancomycin-resistant enterococci in northeast Ohio. Clin Infect Dis 29,573-579[ISI][Medline]
-
Quale, J, Landman, D, Saurina, G, et al (1996) Manipulation of a hospital antimicrobial formulary to control an outbreak of vancomycin-resistant enterococci. Clin Infect Dis 23,1020-1025[ISI][Medline]
-
Paterson DL, Ko WC, Mohapatra S, et al. Klebsiella pneumoniae bacteremia: impact of extended spectrum beta-lactamase (ESBL) production in a global study of 216 patients [abstract]. Presented at 37th Interscience Conference on Antimicrobial Agents and Chemotherapy; September 28 to October 1, 1997; Toronto, Ontario, Canada
-
Lucas, GM, Lechtzin, N, Puryear, DW, et al (1998) Vancomycin-resistant and vancomycin-susceptible enterococcal bacteremia: comparison of clinical features and outcomes. Clin Infect Dis 26,1127-1133[ISI][Medline]
-
Linden, PK, Pasculle, AW, Manez, R, et al (1996) Differences in outcomes for patients with bacteremia due to vancomycin-resistant Enterococcus faecium or vancomycin-susceptible E faecium. Clin Infect Dis 22,663-670[ISI][Medline]
-
Gerding, DN (1997) Is there a relationship between vancomycin-resistant enterococcal infection and Clostridium difficile infection? Clin Infect Dis 25,S206-S210
-
Pallares, R, Pujol, M, Pena, C, et al (1993) Cephalosporins as a risk factor for nosocomial Enterococcus faecalis bacteremia: a matched case-control study. Arch Intern Med 153,1581-1586[Abstract]
-
Boyce, JM, Mermel, LA, Zervos, MJ, et al (1995) Controlling vancomycin-resistant enterococci. Infect Control Hosp Epidemiol 16,634-637[ISI][Medline]
-
Boyce, JM, Opal, SM, Chow, JW, et al (1994) Outbreak of multidrug-resistant Enterococcus faecium with transferable vanB class vancomycin resistance. J Clin Microbiol 32,1148-1153[Abstract/Free Full Text]
-
Morris, JG, Jr, Shay, DK, Hebden, JN, et al (1995) Enterococci resistant to multiple antimicrobial agents, including vancomycin. Ann Intern Med 123,250-259[Abstract/Free Full Text]
-
Bradley, SJ, Wilson, ALT, Allen, MC, et al (1999) The control of hyperendemic glycopeptide-resistant Enterococcus spp. on a haematology unit by changing antibiotic usage. J Antimicrob Chemother 43,261-266[Abstract/Free Full Text]
-
Patterson JE, Przykucki J, Callander N, et al. Control of vancomycin-resistant Enterococcus faecium colonization and infection in a bone marrow transplant unit [abstract]. Presented at 37th annual meeting of the Infectious Disease Society of America; November 18–21, 1999; Philadelphia, PA
This article has been cited by other articles:
|
|
|
|
 
S. N. S. Fonseca, S. R. Melon Kunzle, M. J. Junqueira, R. T. Nascimento, J. I. de Andrade, and A. S. Levin
Implementing 1-Dose Antibiotic Prophylaxis for Prevention of Surgical Site Infection
Arch Surg,
November 1, 2006;
141(11):
1109 - 1113.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
N. Al Naiemi, E. R. Heddema, A. Bart, E. de Jonge, C. M. Vandenbroucke-Grauls, P. H. M. Savelkoul, and B. Duim
Emergence of multidrug-resistant Gram-negative bacteria during selective decontamination of the digestive tract on an intensive care unit
J. Antimicrob. Chemother.,
October 1, 2006;
58(4):
853 - 856.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
N. al Naiemi, A. Bart, M. D. de Jong, C. M. Vandenbroucke-Grauls, P. J. G. M. Rietra, Y. J. Debets-Ossenkopp, P. C. Wever, L. Spanjaard, A. J. Bos, and B. Duim
Widely Distributed and Predominant CTX-M Extended-Spectrum {beta}-Lactamases in Amsterdam, The Netherlands.
J. Clin. Microbiol.,
August 1, 2006;
44(8):
3012 - 3014.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. Mangione-Smith, L. Wong, M. N. Elliott, L. McDonald, and J. Roski
Measuring the Quality of Antibiotic Prescribing for Upper Respiratory Infections and Bronchitis in 5 US Health Plans
Arch Pediatr Adolesc Med,
August 1, 2005;
159(8):
751 - 757.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Nathwani and A. Tice
Ambulatory antimicrobial use: the value of an outcomes registry
J. Antimicrob. Chemother.,
January 1, 2002;
49(1):
149 - 154.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Introduction
