- Email: [email protected]
Strategies for selecting antibiotics in intensive care units
Strategies for selecting antibiotics in intensive care units Clin Micvobiol lnfect 1999; 5: S29-S34
',
Hendrik E. Demey HildeJansens2, Frank Van h e r 2 , Margareta leven3, Herman Goossens3 and Leo L. Bossaert' Departments of 'Intensive Care, 21nfection Control a n d 3Microbiology, University HospitalUniversity of Antwerp UIA, Edegem, Belgium Crowding of severely ill patients in intensive care units has led worldwide to important increases in nosocomial (ICUrelated) infections. Moreover, the nature of these hospital-acquired infections is shifting towards Gram-positive microorganisms, yeast and Gram-negative rods, possessing important resistance genes (e.g. extended spectrum betalactamases and inducible Enterobacteriaceae). Ceftazidime and aztreonam are loosing their activity against the Gramnegative microorganisms. The fourth generation cephalosporins have an intrinsic high activity against the inducible Enterobacteriaceae.On our Hematology and Intensive Care units, the introduction of cefepime for nosocomial infections led to a remarkable drop in the number of Enterobacter isolates combined with important decreases in Enterobacter resistance towards several antibiotics. Key words: Nosocomial infections, antibiotic resistance, fourth generation cephalosporins, cefepime
THE IMPORTANCE OF NOSOCOMIAL INFECTIONS ON THE INTENSIVE CARE UNIT
INTRODUCTION The already busy intensive care physician is faced with an ever increasing problem of bacterial infectioncolonization in the patients hospitalized on intensive care units (ICUs). Prescribing antibiotics is therefore an important task. O f course, other non-antibiotic measures are also needed to prevent or treat these infections [l], such as isolation procedures, hygienic measures, improving the general condition of the patient and, if at all possible, improving inimunity. Many hospitals are currently seeing an increase in the absolute number of nosocomial infections, many of them now multiply resistant especially on intensive care units [2,3], causing increased mortality, length of stay in the I C U or hospital and antibiotic use [4,5].
Concentrating sick people in crowded spaces leads to an increase in the incidence of infections. This was described in the Middle Ages in the pest houses, by Semmelweis stressing the importance of hand washing, by Florence Nightingale and her ideas about planning of hospital facilities [6], and, more recently, by the NNIS study [7,8], the SENIC study [9] and the ICARE project [2]. Modern science has pinpointed the cause of this phenomenon to horizontal transfer of microorganisms between patients, more vulnerable to infection following different invasive procedures and by general decreases in host resistance, due to the steadily increasing age of the population, steroids, chemotherapy for cancer, and immunosuppressive agents. Starting in the 1960s and 1970s the most seriously ill patients were gathered in what we now call intensive care units (ICUs):units where dedicated, highly trained health care workers use complicated machines, invasive procedures and multi-drug cocktails in order to sustain life in victims, for example, of severe trauma, overwhelming infections or cardiovascular crises. The shift
Corresponding author and reprint requests: Dr. H. Derney, University Hospital Antwerp, Department of Intensive Care, Wilrijkstraat 10, 2650 Edegem, Belgium Tel: +32 3 8213635 Fax: +32 3 8284882
E-mail: [email protected] S29
S30
Clinical Microbiology and Infection, Volume 5 Supplement 1
of surgical care to outpatient centers leaves the sickest patients in hospital, which are becoming more like large ICUs and is also leading to a greater prevalence of catheter-related blood stream infections and ventilator-associated pneumonias. Naturally, this crowding of severely ill patients has led to important increases in nosocomial ICU-related infections [10,11]. This was nicely demonstrated by the EPIC study: on 29th April 1992 a one-day prevalence study was carried out in 17 European countries, including more than 10.000 patients in 1,472 ICUs [12]. Of all these patients on modern ICUs, 45%had some form of infection. Thirty per cent were community-acquired, 20% hospital (not 1CU)-acquired and a startling 50% originated on the ICU itself! Of these ICU-acquired infections, two thirds were some form of respiratory infection (mainly colonization?). Staphylococcus aureus and Pseudomonas aeruginosa each were responsible for one third of all infections. Another important factor involved in the rising numbers of nosocomial infections is the health crisis in modern care systems. Several studies suggest a link between decreased nurse staffing levels and an increased risk of infections [13]. The increasing importance of severe infections in patients hospitalized on the ICU, combined with the ever-present threat of nosocomial infection, makes it imperative that the ICU physician has a working knowledge of microbiology, epidemiology, hospital hygiene, clinical infectious diseases and antibiotic therapy. In optimal conditions, he is in daily contact with the microbiological laboratory and with the consultant in infectious diseases. While each patient needs a specific prescription for the treatment of his infection, it is also necessary that a more general set of rules exist for the geographic area (the unit), in which the patient is hospitalized: it has been shown that each geographic area harbors its own indigenous microbial flora, often with important resistance factors. Unfortunately, it has been shown that implementation of standard accepted practices for prevention of hospitalacquired infections is far from satisfactory in European hospitals, even in ICUs [14]. In order to formulate an optimal antibiotic prescription, the physician needs hard data. Sometimes this is available and a choice can be made based on an antibiotic resistance profile. Local prescription guidelines, determined by the hospital antibiotic committee, can help in this decision. Much of the time, certainly on ICU, an antibiotic has to be chosen on empirical grounds: hard data are lacking, or can only be obtained after 12-36 hours. This empirical choice is based on ‘soft’ data. Current microbiological ‘good clinical practice’ is available through several textbooks [I 5-17]. The hospital antibiotic committee guidelines also offer
help. In any case, these soft data lead to a ‘best informed guess’. A combination of data, such as the suspected origin of the infection (e.g. urinary-versus catheterrelated), the most likely organisms involved, the local resistance profile of these microorganisms, and some pharmacokinetic influences (e.g. presence of renal impairment), lead to a logical choice of one or several antibiotics that can be effective in the present clinical condition. Whether a single antibiotic only should be prescribed, or a combination of agents, is still a topic often open to discussion. As the endogenous flora and its potential presence of resistance genes varies between countries, between hospitals and even within the same hospital between specific units, it is necessary to formulate a specific set of prescription rules for each specific unit. In order to do this, the microbiology department must present on a regular basis, an overview of all microorganisms isolated with their sensitivity patterns. The evolutionary changes in the etiology of infections and changes in sensitivity, must be followed and be communicated to the prescribing physicians. This can best be implemented by a coritinuing dialogue between the microbiologist, the antibiotics committee and the end user.
INCREASING ANTIBIOTIC RESISTANCE Such continuing evaluation of the organisms present on ICUs has led to some important conclusions [18]. Increasingly, Gram-positive microorganisms are replacing the Gram-negative rods in invasive infection on the ICU; important differences exist between hospitals in the prevalence of oxacillin resistance. Fungi now occupy fifth place in the rank order of microorganisms isolated from blood cultures; they are also isolated in increasing numbers from respiratory samples. Especially on ICU, inducible Enterobacteriaceae (Enterobacter, Citrobacter, Morganella, and Serratia spp.) present a special problem with increasing resistance towards cephalosporins and quinolones [ 19-2 11. Within the inducible Gram-negatives, Enterobacter spp. species are the most common, causing mainly respiratory colonization and/ or infection. Less frequently they cause blood stream infection or urosepsis. While Pfaller et al. [22] observed clonal spread of a single strain within a given institution, most of the episodes of bacteremia were caused by patient-unique strains of E. cloacae, suggesting that selection of mutant subpopulations within each patient (endogenous infection) by exposure to betalactam agents was an important factor in the development of these resistant strains. In order to combat the rise in antibiotic resistance, different authors have reported on the notion of antibiotic rotation [23]. While the idea is now more than ten years old 1241, no
Derney et al: Strategies f o r s e l e c t i n g a n t i b i o t i c s i n i n t e n s i v e care u n i t s
convincing studies have been published. Other authors stress the fact that, within several antibiotic classes, only individual molecules seem to influence a selective pressure towards resistance [25]. Take ceftazidime, for instance, its use inexorably leads to the selection of resistance in inducible Enterobacteriaceae. O n the other hand, this is not a problem with ceftriaxone. Important also is the fact that quinolone use selects resistance to carbapenems (26,271. During their hospitalization period, the exposure of patients to parenteral antibiotics rose from 23% in 1978 to 44% in 1992, with the average number of different agents used per patient remaining at 1.8-2.1 [28]. Of all the available antibiotics, beta-lactam agents are the most frequently prescribed, especially the cephalosporins (28). In the last decade, ceftazidinie has been the mainstay drug o n many haematology units and ICUs. However, its use is waning for several reasons. Haematology patients present with an increasing number of gram-positive infections, for which ceftazidime is not suited (20% resistance in streptococci, isolated from blood cultures in haematology and oncology patients; MI& 8-32 mg/L in S. aureus). Gram-negative infections have shifted from P aeruginosa and E. coli towards inducible gram-negative infections, for which ceftazidime is losing its activity [19]. Furthermore, extended spectrum beta-lactamases leading to cefiazidinie resistance have appeared, especially in Klebsiella sp. [21].
S31
As in other developed countries, strains isolated from Belgian patients with nosocomial infections, have become more resistant to important antibiotics. An ongoing surveillance study on intensive care (the NPRS study) shows a trend towards progressive increase in resistance of Gram-negative rods; these studies were performed in 1991 [29], 1994-1995 [30] and in 1998 [31] (Figure 1). The problem is especially important for piperacillin, imipeneni, ciprofloxacin and ceftazidime. Aminoglycosides remain active, with only 2 to 5% resistance. Unfortunately, the NPRS study only depicts a trend; absolute changes cannot be calculated due to important changes in methodology between the repetitive studies.
FOURTH GENERATION CEPHALOSPORINS TO THE RESCUE? The fourth generation cephalosporins, cefepime and cefpirome, became available for clinical use at the beginning of the 1990s. Their activity against Grampositive Gicro-organisms is of the same order as cefotaxinie, and much better than ceftazidime. Cefepime is as active against I? aeruginosa as ceftazidime, while cefepime and cefpirome are more active than ceftazidime against inducible Entevobacteriaceae [32]. Reasons for this include the zwitterion structure of the molecules, leading to very rapid membrane penetration and attainment of high concentrations in the periplasmic
-5
IM191
IM195
IM198 CAZ91 CAZ95 CAZ98
PIP91 PIP95 PIP98
PTZ9I PTZ95 PTZ98 AMK91 AMK95 AMK98 CP91
-1
CP95 CP98
Figure 1 Resistance towards different antibiotics for 1991 and 1995 in a sample of Belgian intensive care units from regional and university hospital-the NPRS studies [29-311. SEMC=Inducible Enterobacteriaceae: Servatia, Enterobacter, Movganella and Citrohacter spp. E. coli Kleb=non-inducible Enterobacteriaceae IMI=imipeneni; CAZ=ceftazidime; PIP=piperacillin; PTZ=piperacillin+ tazobactarn; AMK=amikacin; CP=ciprofloxacin.
+
Permission to cite from the NPRS data was obtained from MSD Belgium. The NPRS data are protected by copyright and can only be used, in part or as a whole, after obtaining written permission from Merck Sharpe & Dohnie BV, Waterloosesteenweg 1135, 1180 Brussels, Belgium.
Clinical M i c r o b i o l o g y and Infection, Volume 5 Supplement 1
532
space; high affinity for penicillin binding proteins, good stability against beta-lactamases and low selection pressure for resistance. Cefhirome seems more active against Gram-positives and cefepime more active against the inducible Gram-negatives. Fourth generation cephalosporins could be an important arm in combating the emerging 'epidemic' of ICU-related infections with inducible Gram-negatives, especially Enterobacter sp.
THE ANTWERP EXPERIENCE WITH CEFEPIME IN COMBATING INCREASING ANTIBIOTIC RESISTANCE The University Hospital of Antwerp is a 600-bed hospital, serving as a secondary line and referral hospital. Its potential catchment area includes a population of 1.2 d i o n inhabitants. It offers extensive
cardiovascular surgery services and organ transplant facilities. The ICU consists of 30 fully equipped beds, divided over four adjacent units, each with a specific patient profile. It serves about 2,000 patients per year, with a mean of 4.7 days per admittance. Sixty five per cent of patients are ventilated for more than one day. For several years now, a hospital-wide survey has been performed of all microorganisms isolated according to body site and sensitivity pattern. These data were collected on a departmental basis. Therefore, different guidelines could be created for different clinical units. While the antibiotics committee decided on a hospitalwide antibiotic policy, after discussion it was decided to create a specific subset of prescription rules for the neonatology, hematology and ICU departments, based on the specific needs of these distinct patient classes.
0
E. cloacae E. aerogenes
Numberhonth
J
'M 'M ' J 'S 'N
'
Enterobacter isolates ICU alone
A
15
10
5
0
I
I
I
I
I
I
I
I
I
I
J M M J S N J M M J S N J M M J S N J M M J S N J 'M 'M ' J 'S 'N 1994 1995 1996 1997 1998
'
Figure 2 Incidence of E. aerogenes and E. cloacae on the University Hospital Antwerp Department of Intensive Care. An important increase in incidence in the second half of 1996 led to the introduction of cefepime in the blind 'start' therapy of nosocomial infections from January 1997 onwards. The dramatic fall in Enterobacter isolation is apparent from the graph.
Derney et a l : S t r a t e g i e s for s e l e c t i n g a n t i b i o t i c s in i n t e n s i v e c a r e u n i t s
This ongoing surveillance of the microbiological flora and sensitivity patterns recently led to important guideline changes in both the hematology and ICU department. As reported by Mebis et al. 1331, from May 1995 onwards, ceftazidime monotherapy was replaced by cefepime plus amikacin for empirical therapy of presumed nosocomial infections in hematology patients. Nosocomial infections in the I C U patients were treated with a combination of aztreonam plus cloxacillin plus amikacin until January 1997, when this was switched to cefepime plus amikacin (or cefepime plus ciprofloxacin in a case of failing kidney function). These changes were made, when an important increase in incidence of inducible Enterobacteriaceae was noted in both depart-
Table 1 Percentage resistance of inducible Enterobacteriaceae, isolated from all body sites on the ICU from 1994 onwards. After cefepirne introduction in January 1997, there was a fall in CTZ resistance, while ciprofloxacin resistance kept increasing. Cefepime resistance remained stable over this period Inducible Entrrobacteriaceac all body sites 1994
CTZ AZT
Pw PTZ CIP AMK CEFE IMI
ments, together with increasing resistance towards ceftazidime and aztreonam. In the Hematology department, this change in prescription habits was followed by a sustained decrease in resistance in inducible Enterobacteriaceae towards ceftazidime and ciprofloxacin; the already low resistance in I? aeruginosa remained unchanged [33]. In the intensive care unit, the introduction of cefepime led to a remarkable drop in the number of E. cloacae and, to a lesser extent, E. aerogenes isolates
Table 3 Percentage resistance of inducible Enterobacteriaceae, isolated from respiratory samples on the ICU in 1996 and 1997 Inducible Enterobacteriaceae respiratory samples CTZ AZT PIP PTZ CIP
AMK 1995
1996
1997
17 11 10 13 1 3
39 34 46 33 16 11 11
1
1
55 51 60 50 35 10 8 1
44 43 54 42 42 10 12 9
C T Z =ceftazidime; AZT=aztreonatn; PIP=piperacillin; I’TZ =yiperacilliii-tazobactani; CIP=ciprofloxacin; AMK=arnikaciri; CEFE= ccfepime; IMI =imipeneni.
s33
CEFE IM 1
1996
1997
51 51 54 45 30 7 5
36 36 50 44 39 10 11 6
0
CTZ=ceftazidime; AZT=aztreonam; PIP=pipcraciUiii; PTZ =piperacillin-tazobactam; CIP = ciprofloxacin; AMK =arnikacin; CEFE=crfepime; IMI =imipenern.
(Figure 2). Furthermore, as can be seen from Tables 1-3, resistance towards ceftazidinie fell for isolates of all body sites and ceftazidime resistance in inducible Enterobacteviareae fell in blood and respiratory isolates.
CONCLUSIONS Table 2 Percentage resistance of inducible Enterobacteriaceae, isolated from blood cultures on the ICU from 1994 onwards. Again, there was a fall in ceftazidime and yuinolone resistance. Cefepime resistance remained stable over this period Inducible En terohacteriact-ae blood cultures 1994 CTZ AZT PIP PTZ CII’ AMK CEFE 1MI
14 14 21 21 0 7 0
1995
1996
1997
25 25
38 13
69 s4 69 54 46
64 55 64 55 36
0
8
9
0
8 0
9 10
SO
0
C T Z =ceftazidime; AZT=aztreonatii; PIP= piperacillin; I’TZ =piperacilliii-tazobactam; CIP= ciprofloxacin; AMK=aniikacin; CEFE=cefepime; IMI =imipeneni.
It is necessary to continually evaluate the local microbial flora and its resistance profile in order to formulate specific sets of prescription guidelines and to change these when necessary. Individual antibiotic therapy must always be adapted to the patient and his disease. Inducible Enterobacteriaceae, certainly Enterobactev sp., are at this moment an important problem on the ICU in several countries. Fourth generation cephalosporins possess some important intrinsic molecular advantages which may make them the first choice in settings where empirical therapy is needed. However, continuing evaluation of efficacy is needed, as for other antimicrobial agents. Preliminary data from our hospital’s Heinatology and 1CU departments show that a switch from ceftazidime or aztreonam to cefepime had beneficial effects on infection with inducible Entevobacteriaceae and on resistance profiles.
s34
Clinical Microbiology and Infection, Volume 5 Supplement 1
References 1. Jones S, Fraise A. Coping with nosocomial infection: a nonantibiotic approach. J Hosp Infect 1997; 58: 217-20. 2. Archibald L, Phillips L, Monnet D, McGowann JE, Tenover E Gaynes R. Antimicrobial resistance in isolates from inpatients and outpatients in the United States: increasing importance of the intensive care unit. Clin Infect Dis 1997; 24: 211-5. 3. Spencer R C , Bauernfeind A, Garcia-Rodriguez J, Jarlier V, Pfaller M, Turnidge J et al. Surveillance of the current resistance of nosocomial pathogens to antibacterials. Clin Microbiol Infect 1997; 3(Supplement 1): S21-S35. 4. Girou E, Stephan F, Novara A, Safar M, Fagon J-Y Risk factors and outcome of nosocomial infections: results of a matched casecontrol study of ICU patients. Am J Respir Crit Care Med 1998; 157: 1151-8. 5. Ibelings MMS, Bruining HA. Methicillin-resistant Staphylococcus aureus: acquisition and risk of death in patients in the intensive care unit. Eur J Surg 1998; 164: 411-8. 6. Haley RW. The development of infection surveillance and control programs. In: Bennett JV, Brachman PS (eds). Hospital Infections. 4th ed. Philadelphia and New York: LippincottRaven Publishers; 1998. pp. 5 3 4 4 . 7. Emori T, et al. Nosocomial infections in elderly patients in the United States, 1986-90. Am. J. Med. 1991; 9l(Snppl 3B): 294s. 8. Jarvis W, Edwards J. National Nosocomial Infections Surveillance System. Nosocomial infections in adult and pediatric intensive care units in the United States, 1986-90. Am J Med 1991; 91(Suppl3B): 185s. 9. Haley RW, Culver DH, White JW, Morgan WM, Emori TG, Munn VP et al. The efficacy of infection surveillance and control programs in preventing nosocomial infections in U.S. hospitals. Am J Epidemiol 1985; 121: 182-205. 10. Legras A, Malvy D, Quinioux A, Villers D, Bouachour G, Robert R et al. Nosocomial infections: prospective survey of incidence in five French intensive care units. Int Care Med 1998; 24: 1 0 4 M . 11. Martone WJ, Jarvis WR, Edwards J R , Culver DH, Haley RW. Incidence and nature of endemic and epidemic nosocomial infections. In: Bennett JV, Brachman PS (eds). Hospital Infections, 4th ed. Philadelphia and New York: LippincottRaven Publishers; 1998. pp. 461-7. 12. Vincent JL, Bihari DJ, Suter PM, Bruining HA, White J, Nicolas-Chanoin M H et al. The prevalence of nosocomial infection in intensive care units in Europe. J Am Med Ass 1995; 274: 639-44. 13. McDonald CL, Jarvis WR. Linking antimicrobial use to nosocomid infections: the role of a combined laboratoryepidemiology approach. Ann Int Med 1998; 129: 245-7. 14. Moro M, Jepsen 0,EURO.NIS Study Group. Infection control practices in intensive care units in 14 European countries. Int Care Med 1996; 22: 872-9. 15. Mandell GL, Bennett JE, D o h R (eds). Principles and Practice of Infections Diseases, 4th ed. New York-Edinburgh-LondonMadrid-Melbourne-Milan-Tokyo: Churchill Livingstone; 1995. 16. Keese RE, Betts R F (eds). A Practical Approach to Infectious Diseases. 4th ed. Boston and New York-Toronto-London: Little, Brown and Company; 1996. 17. Gilbert DN, Moellering R C , Jr, Sande MA. The Sanford Guide to Antimicrobial Therapy, 28th ed. Vienna: Antimicrobial Therapy, Inc.; 1998.
18. Wolff M, Brun-Buisson C, Lode H, Mathai D, Lewi D, Pittet D. The changing epidemiology of severe infections in the ICU. Clin Microbiol Infect 1997; 3(Supplement 1): S36S47. 19. Chow J, Fine M, Shlaes D et al. Enterobacter bacteremia: clinical features and emergence of antibiotic resistance during therapy. Ann Intern Med 1991; 115: 585-90. 20. Jacobson K, Cohen S, Inciardi J, King J, Lippert W, Iglesias T et al. The relationship between antecedent antibiotic use and resistance to extended spectrum cephalosporins in group 1 betalactamase producing organisms. Clin Infect Dis 1995; 21: 110713. 21. Rahal JJ, Urban C, Horn D, Freeman K, Segal-Maurer S, Maurer J, et al. Class restriction of cephalosporin use to control total cephalosporin resistance in nosocomial lilebsiefla.J Am Med Ass 1998; 280: 1233-7. 22. Pfaller MA, Jones R N , Marshall SA, C o f h a n SL, Hollis RJ, Edmond MB et al. Inducible Amp C p-lactamase producing gram-negative bacih from blood stream infections: freqency, antimicrobial susceptibility, and molecular epidemiology in a national surveillance program (SCOPE).Diagn Microbiol Infect Dis 1997; 28: 211-9. 23. Niederinan MS. Is ‘crop rotation’ of antibiotics the solution to a ‘resistant’ problem in the ICU? Am J Respir Crit Care Med 1997; 156: 1029-31. 24. McGowan J, Jr. Minimizing antimicrobial resistance in hospital bacteria: Can switching or cycling drugs help? Infect Control 1986; 7: 573. 25. Cunha BA. Antibiotic resistance. Control strategies. Crit Care Clin 1998; 14: 309-27. 26. Kadberg G, Nilssson L, Svensson S. Development ofquinoloneimipenem cross resistance in Pseudomonas aeruginosa during exposure to ciprofloxacin. Antimicrob Agents Chemother 1990; 34: 2142-7. 27. Fung-Tomc J, Kolek B, Bonner D. Ciprofloxacin-induced, lowlevel resistance to structurally unrelated antibiotics in Pseudom o n a ~aeruginosa and methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother 1993; 37: 1289-96. 28. Pallares R , Dick R, Wenzel R, Adams J R , Nettleman MD. Trends in antimicrobial udzation at a tertiary teaching hospital during a 15-year period (1978-1992). Infect Control Hosp Epidemiol 1993; 14: 376-82. 29. Verbist L. Incidence of multi-resistance in Gram-negative bacterial isolates from intensive care units in Belgium: a surveillance study. Scand J Infect Dis 1991 (Supplement 78): 45-53. 30. Glnpczynski Y, Delmte M , Goossens H, Struelens M, Belgian Multicenter ICU Study Group. A multicenter survey of antimicrobial resistance in Gram-negative isolates from Belgian intensive care units in 1994-1995. Acta Clinica Belgica 1998; 53: 28-38. 31. Hanberger H, Garcia-Rodriguz J-A, Gobernado M, Goossens H. Nilsson LE, Strueleiis ME, et al. Antibiotic susceptibility among aerobic Gram-negative bacilli in intensive care units in 5 European countries. JAMA 1999; 281: 67-71. 32. Pierard D, Emmerechts K, Lauwers S, Belgian Multicentre Study Group. Comparative in-vitro activity of cefpirome against isolates from intensive care and haematology/oncology units. J Antimicrob Chemother 1998; 41: 443-50. 33. Mebis J. Goossens H, Bruyneel P, Sion J, Meeus I, Van Droogenbroeck J et al. Decreasing antibiotic resistance ofEnterobacteriaceae by introducing a new antibiotic combination therapy for neutropenic fever patients. Leukemia 1998; 12: 1627-9.
',
Hendrik E. Demey HildeJansens2, Frank Van h e r 2 , Margareta leven3, Herman Goossens3 and Leo L. Bossaert' Departments of 'Intensive Care, 21nfection Control a n d 3Microbiology, University HospitalUniversity of Antwerp UIA, Edegem, Belgium Crowding of severely ill patients in intensive care units has led worldwide to important increases in nosocomial (ICUrelated) infections. Moreover, the nature of these hospital-acquired infections is shifting towards Gram-positive microorganisms, yeast and Gram-negative rods, possessing important resistance genes (e.g. extended spectrum betalactamases and inducible Enterobacteriaceae). Ceftazidime and aztreonam are loosing their activity against the Gramnegative microorganisms. The fourth generation cephalosporins have an intrinsic high activity against the inducible Enterobacteriaceae.On our Hematology and Intensive Care units, the introduction of cefepime for nosocomial infections led to a remarkable drop in the number of Enterobacter isolates combined with important decreases in Enterobacter resistance towards several antibiotics. Key words: Nosocomial infections, antibiotic resistance, fourth generation cephalosporins, cefepime
THE IMPORTANCE OF NOSOCOMIAL INFECTIONS ON THE INTENSIVE CARE UNIT
INTRODUCTION The already busy intensive care physician is faced with an ever increasing problem of bacterial infectioncolonization in the patients hospitalized on intensive care units (ICUs). Prescribing antibiotics is therefore an important task. O f course, other non-antibiotic measures are also needed to prevent or treat these infections [l], such as isolation procedures, hygienic measures, improving the general condition of the patient and, if at all possible, improving inimunity. Many hospitals are currently seeing an increase in the absolute number of nosocomial infections, many of them now multiply resistant especially on intensive care units [2,3], causing increased mortality, length of stay in the I C U or hospital and antibiotic use [4,5].
Concentrating sick people in crowded spaces leads to an increase in the incidence of infections. This was described in the Middle Ages in the pest houses, by Semmelweis stressing the importance of hand washing, by Florence Nightingale and her ideas about planning of hospital facilities [6], and, more recently, by the NNIS study [7,8], the SENIC study [9] and the ICARE project [2]. Modern science has pinpointed the cause of this phenomenon to horizontal transfer of microorganisms between patients, more vulnerable to infection following different invasive procedures and by general decreases in host resistance, due to the steadily increasing age of the population, steroids, chemotherapy for cancer, and immunosuppressive agents. Starting in the 1960s and 1970s the most seriously ill patients were gathered in what we now call intensive care units (ICUs):units where dedicated, highly trained health care workers use complicated machines, invasive procedures and multi-drug cocktails in order to sustain life in victims, for example, of severe trauma, overwhelming infections or cardiovascular crises. The shift
Corresponding author and reprint requests: Dr. H. Derney, University Hospital Antwerp, Department of Intensive Care, Wilrijkstraat 10, 2650 Edegem, Belgium Tel: +32 3 8213635 Fax: +32 3 8284882
E-mail: [email protected] S29
S30
Clinical Microbiology and Infection, Volume 5 Supplement 1
of surgical care to outpatient centers leaves the sickest patients in hospital, which are becoming more like large ICUs and is also leading to a greater prevalence of catheter-related blood stream infections and ventilator-associated pneumonias. Naturally, this crowding of severely ill patients has led to important increases in nosocomial ICU-related infections [10,11]. This was nicely demonstrated by the EPIC study: on 29th April 1992 a one-day prevalence study was carried out in 17 European countries, including more than 10.000 patients in 1,472 ICUs [12]. Of all these patients on modern ICUs, 45%had some form of infection. Thirty per cent were community-acquired, 20% hospital (not 1CU)-acquired and a startling 50% originated on the ICU itself! Of these ICU-acquired infections, two thirds were some form of respiratory infection (mainly colonization?). Staphylococcus aureus and Pseudomonas aeruginosa each were responsible for one third of all infections. Another important factor involved in the rising numbers of nosocomial infections is the health crisis in modern care systems. Several studies suggest a link between decreased nurse staffing levels and an increased risk of infections [13]. The increasing importance of severe infections in patients hospitalized on the ICU, combined with the ever-present threat of nosocomial infection, makes it imperative that the ICU physician has a working knowledge of microbiology, epidemiology, hospital hygiene, clinical infectious diseases and antibiotic therapy. In optimal conditions, he is in daily contact with the microbiological laboratory and with the consultant in infectious diseases. While each patient needs a specific prescription for the treatment of his infection, it is also necessary that a more general set of rules exist for the geographic area (the unit), in which the patient is hospitalized: it has been shown that each geographic area harbors its own indigenous microbial flora, often with important resistance factors. Unfortunately, it has been shown that implementation of standard accepted practices for prevention of hospitalacquired infections is far from satisfactory in European hospitals, even in ICUs [14]. In order to formulate an optimal antibiotic prescription, the physician needs hard data. Sometimes this is available and a choice can be made based on an antibiotic resistance profile. Local prescription guidelines, determined by the hospital antibiotic committee, can help in this decision. Much of the time, certainly on ICU, an antibiotic has to be chosen on empirical grounds: hard data are lacking, or can only be obtained after 12-36 hours. This empirical choice is based on ‘soft’ data. Current microbiological ‘good clinical practice’ is available through several textbooks [I 5-17]. The hospital antibiotic committee guidelines also offer
help. In any case, these soft data lead to a ‘best informed guess’. A combination of data, such as the suspected origin of the infection (e.g. urinary-versus catheterrelated), the most likely organisms involved, the local resistance profile of these microorganisms, and some pharmacokinetic influences (e.g. presence of renal impairment), lead to a logical choice of one or several antibiotics that can be effective in the present clinical condition. Whether a single antibiotic only should be prescribed, or a combination of agents, is still a topic often open to discussion. As the endogenous flora and its potential presence of resistance genes varies between countries, between hospitals and even within the same hospital between specific units, it is necessary to formulate a specific set of prescription rules for each specific unit. In order to do this, the microbiology department must present on a regular basis, an overview of all microorganisms isolated with their sensitivity patterns. The evolutionary changes in the etiology of infections and changes in sensitivity, must be followed and be communicated to the prescribing physicians. This can best be implemented by a coritinuing dialogue between the microbiologist, the antibiotics committee and the end user.
INCREASING ANTIBIOTIC RESISTANCE Such continuing evaluation of the organisms present on ICUs has led to some important conclusions [18]. Increasingly, Gram-positive microorganisms are replacing the Gram-negative rods in invasive infection on the ICU; important differences exist between hospitals in the prevalence of oxacillin resistance. Fungi now occupy fifth place in the rank order of microorganisms isolated from blood cultures; they are also isolated in increasing numbers from respiratory samples. Especially on ICU, inducible Enterobacteriaceae (Enterobacter, Citrobacter, Morganella, and Serratia spp.) present a special problem with increasing resistance towards cephalosporins and quinolones [ 19-2 11. Within the inducible Gram-negatives, Enterobacter spp. species are the most common, causing mainly respiratory colonization and/ or infection. Less frequently they cause blood stream infection or urosepsis. While Pfaller et al. [22] observed clonal spread of a single strain within a given institution, most of the episodes of bacteremia were caused by patient-unique strains of E. cloacae, suggesting that selection of mutant subpopulations within each patient (endogenous infection) by exposure to betalactam agents was an important factor in the development of these resistant strains. In order to combat the rise in antibiotic resistance, different authors have reported on the notion of antibiotic rotation [23]. While the idea is now more than ten years old 1241, no
Derney et al: Strategies f o r s e l e c t i n g a n t i b i o t i c s i n i n t e n s i v e care u n i t s
convincing studies have been published. Other authors stress the fact that, within several antibiotic classes, only individual molecules seem to influence a selective pressure towards resistance [25]. Take ceftazidime, for instance, its use inexorably leads to the selection of resistance in inducible Enterobacteriaceae. O n the other hand, this is not a problem with ceftriaxone. Important also is the fact that quinolone use selects resistance to carbapenems (26,271. During their hospitalization period, the exposure of patients to parenteral antibiotics rose from 23% in 1978 to 44% in 1992, with the average number of different agents used per patient remaining at 1.8-2.1 [28]. Of all the available antibiotics, beta-lactam agents are the most frequently prescribed, especially the cephalosporins (28). In the last decade, ceftazidinie has been the mainstay drug o n many haematology units and ICUs. However, its use is waning for several reasons. Haematology patients present with an increasing number of gram-positive infections, for which ceftazidime is not suited (20% resistance in streptococci, isolated from blood cultures in haematology and oncology patients; MI& 8-32 mg/L in S. aureus). Gram-negative infections have shifted from P aeruginosa and E. coli towards inducible gram-negative infections, for which ceftazidime is losing its activity [19]. Furthermore, extended spectrum beta-lactamases leading to cefiazidinie resistance have appeared, especially in Klebsiella sp. [21].
S31
As in other developed countries, strains isolated from Belgian patients with nosocomial infections, have become more resistant to important antibiotics. An ongoing surveillance study on intensive care (the NPRS study) shows a trend towards progressive increase in resistance of Gram-negative rods; these studies were performed in 1991 [29], 1994-1995 [30] and in 1998 [31] (Figure 1). The problem is especially important for piperacillin, imipeneni, ciprofloxacin and ceftazidime. Aminoglycosides remain active, with only 2 to 5% resistance. Unfortunately, the NPRS study only depicts a trend; absolute changes cannot be calculated due to important changes in methodology between the repetitive studies.
FOURTH GENERATION CEPHALOSPORINS TO THE RESCUE? The fourth generation cephalosporins, cefepime and cefpirome, became available for clinical use at the beginning of the 1990s. Their activity against Grampositive Gicro-organisms is of the same order as cefotaxinie, and much better than ceftazidime. Cefepime is as active against I? aeruginosa as ceftazidime, while cefepime and cefpirome are more active than ceftazidime against inducible Entevobacteriaceae [32]. Reasons for this include the zwitterion structure of the molecules, leading to very rapid membrane penetration and attainment of high concentrations in the periplasmic
-5
IM191
IM195
IM198 CAZ91 CAZ95 CAZ98
PIP91 PIP95 PIP98
PTZ9I PTZ95 PTZ98 AMK91 AMK95 AMK98 CP91
-1
CP95 CP98
Figure 1 Resistance towards different antibiotics for 1991 and 1995 in a sample of Belgian intensive care units from regional and university hospital-the NPRS studies [29-311. SEMC=Inducible Enterobacteriaceae: Servatia, Enterobacter, Movganella and Citrohacter spp. E. coli Kleb=non-inducible Enterobacteriaceae IMI=imipeneni; CAZ=ceftazidime; PIP=piperacillin; PTZ=piperacillin+ tazobactarn; AMK=amikacin; CP=ciprofloxacin.
+
Permission to cite from the NPRS data was obtained from MSD Belgium. The NPRS data are protected by copyright and can only be used, in part or as a whole, after obtaining written permission from Merck Sharpe & Dohnie BV, Waterloosesteenweg 1135, 1180 Brussels, Belgium.
Clinical M i c r o b i o l o g y and Infection, Volume 5 Supplement 1
532
space; high affinity for penicillin binding proteins, good stability against beta-lactamases and low selection pressure for resistance. Cefhirome seems more active against Gram-positives and cefepime more active against the inducible Gram-negatives. Fourth generation cephalosporins could be an important arm in combating the emerging 'epidemic' of ICU-related infections with inducible Gram-negatives, especially Enterobacter sp.
THE ANTWERP EXPERIENCE WITH CEFEPIME IN COMBATING INCREASING ANTIBIOTIC RESISTANCE The University Hospital of Antwerp is a 600-bed hospital, serving as a secondary line and referral hospital. Its potential catchment area includes a population of 1.2 d i o n inhabitants. It offers extensive
cardiovascular surgery services and organ transplant facilities. The ICU consists of 30 fully equipped beds, divided over four adjacent units, each with a specific patient profile. It serves about 2,000 patients per year, with a mean of 4.7 days per admittance. Sixty five per cent of patients are ventilated for more than one day. For several years now, a hospital-wide survey has been performed of all microorganisms isolated according to body site and sensitivity pattern. These data were collected on a departmental basis. Therefore, different guidelines could be created for different clinical units. While the antibiotics committee decided on a hospitalwide antibiotic policy, after discussion it was decided to create a specific subset of prescription rules for the neonatology, hematology and ICU departments, based on the specific needs of these distinct patient classes.
0
E. cloacae E. aerogenes
Numberhonth
J
'M 'M ' J 'S 'N
'
Enterobacter isolates ICU alone
A
15
10
5
0
I
I
I
I
I
I
I
I
I
I
J M M J S N J M M J S N J M M J S N J M M J S N J 'M 'M ' J 'S 'N 1994 1995 1996 1997 1998
'
Figure 2 Incidence of E. aerogenes and E. cloacae on the University Hospital Antwerp Department of Intensive Care. An important increase in incidence in the second half of 1996 led to the introduction of cefepime in the blind 'start' therapy of nosocomial infections from January 1997 onwards. The dramatic fall in Enterobacter isolation is apparent from the graph.
Derney et a l : S t r a t e g i e s for s e l e c t i n g a n t i b i o t i c s in i n t e n s i v e c a r e u n i t s
This ongoing surveillance of the microbiological flora and sensitivity patterns recently led to important guideline changes in both the hematology and ICU department. As reported by Mebis et al. 1331, from May 1995 onwards, ceftazidime monotherapy was replaced by cefepime plus amikacin for empirical therapy of presumed nosocomial infections in hematology patients. Nosocomial infections in the I C U patients were treated with a combination of aztreonam plus cloxacillin plus amikacin until January 1997, when this was switched to cefepime plus amikacin (or cefepime plus ciprofloxacin in a case of failing kidney function). These changes were made, when an important increase in incidence of inducible Enterobacteriaceae was noted in both depart-
Table 1 Percentage resistance of inducible Enterobacteriaceae, isolated from all body sites on the ICU from 1994 onwards. After cefepirne introduction in January 1997, there was a fall in CTZ resistance, while ciprofloxacin resistance kept increasing. Cefepime resistance remained stable over this period Inducible Entrrobacteriaceac all body sites 1994
CTZ AZT
Pw PTZ CIP AMK CEFE IMI
ments, together with increasing resistance towards ceftazidime and aztreonam. In the Hematology department, this change in prescription habits was followed by a sustained decrease in resistance in inducible Enterobacteriaceae towards ceftazidime and ciprofloxacin; the already low resistance in I? aeruginosa remained unchanged [33]. In the intensive care unit, the introduction of cefepime led to a remarkable drop in the number of E. cloacae and, to a lesser extent, E. aerogenes isolates
Table 3 Percentage resistance of inducible Enterobacteriaceae, isolated from respiratory samples on the ICU in 1996 and 1997 Inducible Enterobacteriaceae respiratory samples CTZ AZT PIP PTZ CIP
AMK 1995
1996
1997
17 11 10 13 1 3
39 34 46 33 16 11 11
1
1
55 51 60 50 35 10 8 1
44 43 54 42 42 10 12 9
C T Z =ceftazidime; AZT=aztreonatn; PIP=piperacillin; I’TZ =yiperacilliii-tazobactani; CIP=ciprofloxacin; AMK=arnikaciri; CEFE= ccfepime; IMI =imipeneni.
s33
CEFE IM 1
1996
1997
51 51 54 45 30 7 5
36 36 50 44 39 10 11 6
0
CTZ=ceftazidime; AZT=aztreonam; PIP=pipcraciUiii; PTZ =piperacillin-tazobactam; CIP = ciprofloxacin; AMK =arnikacin; CEFE=crfepime; IMI =imipenern.
(Figure 2). Furthermore, as can be seen from Tables 1-3, resistance towards ceftazidinie fell for isolates of all body sites and ceftazidime resistance in inducible Enterobacteviareae fell in blood and respiratory isolates.
CONCLUSIONS Table 2 Percentage resistance of inducible Enterobacteriaceae, isolated from blood cultures on the ICU from 1994 onwards. Again, there was a fall in ceftazidime and yuinolone resistance. Cefepime resistance remained stable over this period Inducible En terohacteriact-ae blood cultures 1994 CTZ AZT PIP PTZ CII’ AMK CEFE 1MI
14 14 21 21 0 7 0
1995
1996
1997
25 25
38 13
69 s4 69 54 46
64 55 64 55 36
0
8
9
0
8 0
9 10
SO
0
C T Z =ceftazidime; AZT=aztreonatii; PIP= piperacillin; I’TZ =piperacilliii-tazobactam; CIP= ciprofloxacin; AMK=aniikacin; CEFE=cefepime; IMI =imipeneni.
It is necessary to continually evaluate the local microbial flora and its resistance profile in order to formulate specific sets of prescription guidelines and to change these when necessary. Individual antibiotic therapy must always be adapted to the patient and his disease. Inducible Enterobacteriaceae, certainly Enterobactev sp., are at this moment an important problem on the ICU in several countries. Fourth generation cephalosporins possess some important intrinsic molecular advantages which may make them the first choice in settings where empirical therapy is needed. However, continuing evaluation of efficacy is needed, as for other antimicrobial agents. Preliminary data from our hospital’s Heinatology and 1CU departments show that a switch from ceftazidime or aztreonam to cefepime had beneficial effects on infection with inducible Entevobacteriaceae and on resistance profiles.
s34
Clinical Microbiology and Infection, Volume 5 Supplement 1
References 1. Jones S, Fraise A. Coping with nosocomial infection: a nonantibiotic approach. J Hosp Infect 1997; 58: 217-20. 2. Archibald L, Phillips L, Monnet D, McGowann JE, Tenover E Gaynes R. Antimicrobial resistance in isolates from inpatients and outpatients in the United States: increasing importance of the intensive care unit. Clin Infect Dis 1997; 24: 211-5. 3. Spencer R C , Bauernfeind A, Garcia-Rodriguez J, Jarlier V, Pfaller M, Turnidge J et al. Surveillance of the current resistance of nosocomial pathogens to antibacterials. Clin Microbiol Infect 1997; 3(Supplement 1): S21-S35. 4. Girou E, Stephan F, Novara A, Safar M, Fagon J-Y Risk factors and outcome of nosocomial infections: results of a matched casecontrol study of ICU patients. Am J Respir Crit Care Med 1998; 157: 1151-8. 5. Ibelings MMS, Bruining HA. Methicillin-resistant Staphylococcus aureus: acquisition and risk of death in patients in the intensive care unit. Eur J Surg 1998; 164: 411-8. 6. Haley RW. The development of infection surveillance and control programs. In: Bennett JV, Brachman PS (eds). Hospital Infections. 4th ed. Philadelphia and New York: LippincottRaven Publishers; 1998. pp. 5 3 4 4 . 7. Emori T, et al. Nosocomial infections in elderly patients in the United States, 1986-90. Am. J. Med. 1991; 9l(Snppl 3B): 294s. 8. Jarvis W, Edwards J. National Nosocomial Infections Surveillance System. Nosocomial infections in adult and pediatric intensive care units in the United States, 1986-90. Am J Med 1991; 91(Suppl3B): 185s. 9. Haley RW, Culver DH, White JW, Morgan WM, Emori TG, Munn VP et al. The efficacy of infection surveillance and control programs in preventing nosocomial infections in U.S. hospitals. Am J Epidemiol 1985; 121: 182-205. 10. Legras A, Malvy D, Quinioux A, Villers D, Bouachour G, Robert R et al. Nosocomial infections: prospective survey of incidence in five French intensive care units. Int Care Med 1998; 24: 1 0 4 M . 11. Martone WJ, Jarvis WR, Edwards J R , Culver DH, Haley RW. Incidence and nature of endemic and epidemic nosocomial infections. In: Bennett JV, Brachman PS (eds). Hospital Infections, 4th ed. Philadelphia and New York: LippincottRaven Publishers; 1998. pp. 461-7. 12. Vincent JL, Bihari DJ, Suter PM, Bruining HA, White J, Nicolas-Chanoin M H et al. The prevalence of nosocomial infection in intensive care units in Europe. J Am Med Ass 1995; 274: 639-44. 13. McDonald CL, Jarvis WR. Linking antimicrobial use to nosocomid infections: the role of a combined laboratoryepidemiology approach. Ann Int Med 1998; 129: 245-7. 14. Moro M, Jepsen 0,EURO.NIS Study Group. Infection control practices in intensive care units in 14 European countries. Int Care Med 1996; 22: 872-9. 15. Mandell GL, Bennett JE, D o h R (eds). Principles and Practice of Infections Diseases, 4th ed. New York-Edinburgh-LondonMadrid-Melbourne-Milan-Tokyo: Churchill Livingstone; 1995. 16. Keese RE, Betts R F (eds). A Practical Approach to Infectious Diseases. 4th ed. Boston and New York-Toronto-London: Little, Brown and Company; 1996. 17. Gilbert DN, Moellering R C , Jr, Sande MA. The Sanford Guide to Antimicrobial Therapy, 28th ed. Vienna: Antimicrobial Therapy, Inc.; 1998.
18. Wolff M, Brun-Buisson C, Lode H, Mathai D, Lewi D, Pittet D. The changing epidemiology of severe infections in the ICU. Clin Microbiol Infect 1997; 3(Supplement 1): S36S47. 19. Chow J, Fine M, Shlaes D et al. Enterobacter bacteremia: clinical features and emergence of antibiotic resistance during therapy. Ann Intern Med 1991; 115: 585-90. 20. Jacobson K, Cohen S, Inciardi J, King J, Lippert W, Iglesias T et al. The relationship between antecedent antibiotic use and resistance to extended spectrum cephalosporins in group 1 betalactamase producing organisms. Clin Infect Dis 1995; 21: 110713. 21. Rahal JJ, Urban C, Horn D, Freeman K, Segal-Maurer S, Maurer J, et al. Class restriction of cephalosporin use to control total cephalosporin resistance in nosocomial lilebsiefla.J Am Med Ass 1998; 280: 1233-7. 22. Pfaller MA, Jones R N , Marshall SA, C o f h a n SL, Hollis RJ, Edmond MB et al. Inducible Amp C p-lactamase producing gram-negative bacih from blood stream infections: freqency, antimicrobial susceptibility, and molecular epidemiology in a national surveillance program (SCOPE).Diagn Microbiol Infect Dis 1997; 28: 211-9. 23. Niederinan MS. Is ‘crop rotation’ of antibiotics the solution to a ‘resistant’ problem in the ICU? Am J Respir Crit Care Med 1997; 156: 1029-31. 24. McGowan J, Jr. Minimizing antimicrobial resistance in hospital bacteria: Can switching or cycling drugs help? Infect Control 1986; 7: 573. 25. Cunha BA. Antibiotic resistance. Control strategies. Crit Care Clin 1998; 14: 309-27. 26. Kadberg G, Nilssson L, Svensson S. Development ofquinoloneimipenem cross resistance in Pseudomonas aeruginosa during exposure to ciprofloxacin. Antimicrob Agents Chemother 1990; 34: 2142-7. 27. Fung-Tomc J, Kolek B, Bonner D. Ciprofloxacin-induced, lowlevel resistance to structurally unrelated antibiotics in Pseudom o n a ~aeruginosa and methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother 1993; 37: 1289-96. 28. Pallares R , Dick R, Wenzel R, Adams J R , Nettleman MD. Trends in antimicrobial udzation at a tertiary teaching hospital during a 15-year period (1978-1992). Infect Control Hosp Epidemiol 1993; 14: 376-82. 29. Verbist L. Incidence of multi-resistance in Gram-negative bacterial isolates from intensive care units in Belgium: a surveillance study. Scand J Infect Dis 1991 (Supplement 78): 45-53. 30. Glnpczynski Y, Delmte M , Goossens H, Struelens M, Belgian Multicenter ICU Study Group. A multicenter survey of antimicrobial resistance in Gram-negative isolates from Belgian intensive care units in 1994-1995. Acta Clinica Belgica 1998; 53: 28-38. 31. Hanberger H, Garcia-Rodriguz J-A, Gobernado M, Goossens H. Nilsson LE, Strueleiis ME, et al. Antibiotic susceptibility among aerobic Gram-negative bacilli in intensive care units in 5 European countries. JAMA 1999; 281: 67-71. 32. Pierard D, Emmerechts K, Lauwers S, Belgian Multicentre Study Group. Comparative in-vitro activity of cefpirome against isolates from intensive care and haematology/oncology units. J Antimicrob Chemother 1998; 41: 443-50. 33. Mebis J. Goossens H, Bruyneel P, Sion J, Meeus I, Van Droogenbroeck J et al. Decreasing antibiotic resistance ofEnterobacteriaceae by introducing a new antibiotic combination therapy for neutropenic fever patients. Leukemia 1998; 12: 1627-9.