SciELO - Scientific Electronic Library Online

 
vol.28 número1Mecanismos del dolor neuropático: Del laboratorio a la clínicaEstudio de eficacia clínica y calidad de anestésicos locales usados en la práctica odontológica índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados

Revista

Articulo

Indicadores

Links relacionados

  • No hay articulos similaresSimilares en SciELO

Compartir


Archivos Venezolanos de Farmacología y Terapéutica

versión impresa ISSN 0798-0264

AVFT v.28 n.1 Caracas ene. 2009

 

A prospective microbial surveillance project at a major Italian Hospital. Epidemiological perspectives, and in vitro susceptibility rates.

Running headline: Microbiological surveillance, antibiotic resistance monitoring

Roberto Manfredi, MD

Department of Internal Medicine, Aging, and Nephrologic Diseases, Division of Infectious Diseases, “Alma Mater Studiorum” University of Bologna, S. Orsola-Malpighi Hospital, Bologna, Italy.

Conflicts of interest, competing interests, funding, sponsorship, acknowledgement: none Correspondence: Prof. Roberto Manfredi Infectious Diseases, University of Bologna S. Orsola Hospital, Via Massarenti 11. 40138 Bologna, Italy Telephone: 051-6363355 Telefax: 051-343500 E-mail: Roberto.manfredi@unibo.it

Summary

Objective. Microbiological surveillance program is currently performed at our tertiary-care teaching Hospital. Patients and Methods. The temporal trend of microbial isolates from patients admitted during the last four calendar years (2004 to 2007), has been analyzed according to the main bacterial and fungal cultured organisms. The same pathogens isolated more than once from the same patient within one month, have been considered only once. Results. On the whole, the main pathogen group remained that of Enterobacteriaceae (6,608 isolations out of 19,666: 33.6%, with Escherichia coli retrieved in 60-75% of cases), with no significant difference over time. Staphylococci (4,150 isolates), and enterococci (3,276 isolates), were the two largest groups after Enterobacteriaceae, but staphylococci significantly declined during the examined four-year period (p<.001), mainly due to a progressively reduced isolation of coagulase-negative staphylococci. On the other hand, a slight increase of enterococci occurred (p<.05). Based on the frequency of isolation, Gram-negative oxydasepositive organisms accounted for 2,109 episodes, followed by other aerobe Gram-positive organisms other than Staphylococci-Enterococci (613 isolates), and anaerobes (583 isolates): no significant temporal variations occurred over time for these last microbial groups. With regard to Gram-negative oxydase-negative microorganisms (567 isolates), non-betahemolytic streptococci (464 cases), and beta-hemolytic streptococci (260 isolates), a significant trend towards a reduction of frequency occurred from the year 2004 to 2007 (p<.05 to p<.001). Finally, fungal infections accounted for 1,036 overall episodes, in over 80% of cases represented by Candida spp. Discussion. A prospective microbiological monitoring is expected to significantly contribute to the knowledge of local epidemiological figures and antimicrobial sensitivity profile of hospital infections, and plays a relevant role in the selection of both treatment and chemoprophylaxis schedules, especially on local-regional basis. Although the major causative agents of inpatient infections remain Enterobacteriaceae, however a significant decline of coagulase-negative Staphylococci, all Streptococci, and Gram-negative oxydase-negative organisms occurred over the examined four-year period, while Enterococci showed a mild increase over time.

Key words: Microbiological surveillance, hospital, microbial isolates, epidemiological trends, in vitro antimicrobial susceptibility trends, clinical concerns

Recibido: 06/01/2009 Aceptado: 04/03/2009

Introduction

Novel epidemilogical changes and resistance problems have emerged recently, especially among hospital pathogens like Staphylococci, Enterococci, Pseudomonas spp., and other Gram-negative rods. Whereas novel molecules have been and are being developed for treating Gram-positive infections, difficult pandrug-resistant Gram-negative infections are expected to become an increasing therapeutic challenge in the next future months1-5.

Routine hospital (and possibly communty) surveillance programs are critical for creating and refining approaching to controlling antimicrobial resistance, and for guiding clinical decisions on the pathway of appropriate drug choices. Longitudinal studies are particularly useful since relevant informations can be obtained by confronting data over time. In our case, an active, prospective microbiological surveillance of all bacterial and fungal microorganisms isolated from inpatients at a tertiary care reference hospital, and their in vitro antimicrobial susceptibility trends, on our opinion is of paramount importance in building awareness of evolving local epidemiology, and in planning effective preventive and therapeutic strategies3,4,6.

Moreover, the knowledge of the clinical and economic impact of multiresistant infections and antimicrobial resistance is extremely useful to influence programs and behavior in health care facilities, to guide policy makers and funding institutions in the light of length of hospital stay, adverse events, and increased expenditures, to define the outcome of individual patients, and to stimulate interest in a correct use and especially in the research and development of new antiinfective agents2-5,7.

Patients and methods

An active bacteriological-mycological surveillance project has been judged mandatory as a part of the necessary basic knowledge on the local microbial isolation, and the local microorganism and antibiotic resistance maps, at our Hospital (S. Orsola-Malpighi General Hospital, Bologna, Italy).

Complete figures and trends of microbial (bacterial and fungal) isolations from all hospitalized patients are consistently registered in a standardized software, and are periodically reported to all health care providers on quarterly basis, together with results of the in vitro antimicrobial sensitivity testing8, compared with those of previous months and years.

For the present report, a four-year period (rangimg from January 1, 2004, up to December 31, 2007), has been comprehensively assessed to observe eventual epidemiological trends, to focus several emerging clinical problems, and to evaluate the parallel trend of antimicrobial sensitivity assays systematically performed on the isolated organisms.

Some clinical and microbiological topics (i.e. central vascular catheter infections, urinary tract infections, in vitro susceptibility of emerging microorganisms like Enterococci and Pseudomonas aeruginosa, as well as the phenomenon of extended spectrum beta-lactamase (ESBL) production, have been analyzed in depth, on the ground of the updated data of the last available year of observation (year 2007).

Commercial, standardized laboratory techniques were consistently used during the entire study period, in order to obtain microbial growth and reliable identification, and to assess the in vitro antimicrobial susceptibility rates according to the updated international reference standards8.

From a methodological point of view, all surveillance cultures were ruled out from the present study, while multiple isolations of the same organism(s) obtained from one single patient within one month of hospitalization, accounted for one isolation only (i.e., they have been considered only once), in order to reduce isolation biases, and their subsequent microbiological and antimicrobial susceptibility analysis, performed on overall basis.From a statistical point of view, continuous variables were assessed through Student t test, while Mantel-Haenzel chisquare test (or Fisher exact test), were adopted to evaluate non-continuous variables. Statistical significance has been posed at p values <.05, as usual.

Results

The global observation period (2004-2007) The temporal trend of microbial isolates from patients admitted during the last four calendar years (2004 to 2007), has been assessed according to the main bacterial and fungal cultures obtained from inpatients, as specified above.

On the whole, the main pathogen group remained that of Enterobacteriaceae (6,608 isolations out of 19,666 overall strains: equivalent to 33.6%, with Escherichia coli retrieved in 60-75% of cases), without significant differences noticed over the four-year observation time. Staphylococci (4,150 isolates), and Enterococci (3,276 isolates), proved the two largest groups after Enterobacteriaceae, but Staphylococci significantly declined in their frequency during the examined four-year period (from 23.8% to 16.2%; p<.001), mainly due to a reduced isolation of coagulase-negative Staphylococci (possibly attributable to a reduced rate of contaminated blood cultures). In the same time, a slight increase of Enterococci occurred (from 15.4% of year 2004 to 18.6% of year 2007; p<.05). Based on the frequency of isolation, Gram-negative oxydase-positive organisms accounted for 2,109 episodes, followed by other aerobe Gram-positive organisms other than Staphylococci-Enterococci (613 isolates), and anaerobe organisms (583 isolates): no significant temporal variations occurred over time for these last microbial groups. With regard to Gram-negative oxydase-negative microorganisms (567 isolates), non-beta-hemolytic Streptococci (464 cases), and beta-hemolytic Streptococci (260 isolates), a significant trend towards a reduction of frequency occurred from the year 2004 to 2007 (p<.05 to p<.001). Finally, fungal infections accounted for 1,036 overall episodes, in over 80% of cases represented by the yeast Candida spp.

Updated microbiology: year 2007

On the whole, 5,256 overall evaluable pathogens were cultured and identified in the last 12 evaluable months (January 1, up to December 31, 2007), with Eschericha coli (1,144 strains), Enterococcus faecalis (697), Pseudomonas aeruginosa (439), and Staphylococcus aureus (389), as the prevailing organisms.

Among blood cultures (1,035 overall isolates), a major role was played by Staphylococci as a group (328), followed by Enterobacteriaceae (222), and Enterococci (114).

With regard to the overall susceptibility rates, methicillin-resistant S. aureus accounted for a mean 46.5% rate, while methicillin resistance was greater for S. epidermidis (83.5%); anyway, no significant changes were found during the examined time. Substantially unchanged in vitro sensitivity levels were found among Enterococci, with only four “Van-A” E. faecalis strains, and 14 “Van-B” E. faecium strains recorded: a slight increase of resistance towards glycopeptides was therefore found, reaching nearly 3% of isolates, concetrated in the last three months of the year 2007. A 26-40% resistance rate to macrolides was found among Streptococci, without appreciable temporal variations. Among Gram-negative organisms, extended spectrum beta-lactamase (ESBL) production regarded 23.5% of Escherichia coli isolates, and 38.3% of other Enterobacteriaceae, again without a significant increase during time. P. aeruginosa showed an increased antiobitic resistance trend towards penicillins and cephalosporins (including those protected by beta-lactamase inhibitors), carbapanems, fluoroquinolones, and aminoglycosides (60-80% of strains), but remained full susceptible to colistin. Stenotrophomonas maltophilia confirmed its extensive resistance spectrum, but remained 90-100% sensitive to cotrimoxazole and colistin, as well as Acinetobacter spp., which showed a favorable susceptibility rate (60-100%) to aminoglycosides, piperacillintazobactam, and colistin.

Focus on sepsis-bacteremia (2004-2007)

Of 4,606 overall episodes of hospital-diagnosed sepsis-bacteremia, S. epidermidis remained the leading organism (983 cases: 21.3%), but a dramatic drop in its frequency occurred during the observation time (from 26.1% of overall episodes in the year 2004, to 14.3% only in the year 2007; p<.0001). The second cause of bacteremia was represented by E. coli (463 episodes: 10.1%), followed by S. aureus (327 cases: 7.1%), Enterococcus faecalis (245 episodes: 5.3%), Pseudomonas aeruginosa (170 cases: 3.7%), Klebsiella spp. (123 episodes: 2.7%), and Enterococcus faecium (117 cases: 2.5%). Significant time-based modifications occurred only for P. aeruginosa (temporal increase: p<.02), Klebsiella spp. (temporal increase: p<.001), and E. faecium (temporal increase: p<.05). Among fungi, Candida albicans was the most represented organism, with 104 episodes (2.3%), without changes in its frequency in the examined 2004-2007 period.

Focus on central vascular catheter infections. Update of the year 2007

The trend of microbial isolations from patients admitted during the last calendar year (January 1, to December 31, 2007), with a clinically- and microbiologically-confirmed central venous catheter (CVC) infection, was regularly reported on quarterly basis. The trend of CVC infections monitorized among our inpatients moderately varied during the observation period (149 cases in January-March, 169 episodes in April-June, 129 cases in July-September, and 134 episodes in October-December). Among the most frequent organisms, S. epidermidis accounted for the majority of isolates (183 cases: 31.5%), followed by E. coli (49: 8.4%), S. aureus (45: 7.7%), P. aeruginosa (36: 6.2%), E. faecalis (30: 5.2%), E. faecium (25: 4.3%), Klebsiella pneumoniae (21: 3.6%), and Enterobacter cloacae (15: 2.6%), while the yeast Candida albicans accounted for a minority of episodes (17 cases only: 2.9%). When analyzing the available figures according to calendar months of year 2007, only some Gram-negative pathogens showed an increased incidence over time: P. aeruginosa from 5.4% in the first three months of 2007 up to 7.5% in the last three months of 2007, and E. cloacae (from 2.0% in January- March 2007, up to 2.68% in October-December 2007), as well as other environmental Gram-negative organisms.

Focus on urinary tract infections. Update of the year 2007

The temporal trend of microbial isolates from urines of inpatients hospitalized during the last calendar year (2007), was evaluated quarterly according to the main bacterial and fungal isolates. Of 2.384 overall episodes (as defined above), 920 (38.6%) were determined by E. coli, followed by E. faecalis (541 episodes: 22.7%), P. aeruginosa (140 cases: 5.9%), Klebsiella spp. (134 episodes: 5.6%), Proteus mirabilis (118 cases: 4.9%), E. faecium (76 episodes: 3.2%), while the first fungal pathogen was again Candida albicans (87 episodes: 3.6%). During the proportionally short (one-year) observation period, no significant variations occurred in the frequency of isolation of each mentioned urinary tract pathogen, when excluding a mild increased frequency of Enterococcus faecium during time (p<.05).

A four-year trend of in vitro sensitivity profile of Staphylococcus aureus strains

The progressively increased rate of drug resistance among Gram-positive cocci is a general concern, especially in hospital settings. The temporal variations of the in vitro antimicrobial sensitivity figures were examined quarterly for all suitable S. aureus strains, and followed from year 2004 to year 2007. Among overall S. aureus isolates (1.863 strains tested on the whole), a complete (100%) sensitivity was shown against both glycopeptide antibiotics vancomycin and teicoplanin, while some compounds retained interesting activity (92.0% to 97.1% for cotrimoxazole, 76.1% to 88.7% for chloramphenicol, 64.1% to 69.5% for rifampicin). Oxacillin (methicillin) resistance ranged from a minimum mean rate of 46.2% (year 2007), to a maximum mean rate of 53.3%, registered during the year 2005. As a consequence, beta-lactam derivatives proved an in vitro activity ranging from 46.7% to 54.1% for co-amoxiclav, 46.6% to 54.1% for cefotaxime, and only 7.3% to 11.2% for penicillin. Among other tested molecules, clindamycin reached a comprehensive ~50% susceptibility rate (40.8% to 54.3% of all tested strains), followed by erythromycin (40.3% to 54.7%), and gentamicin (42.6% to 49-5%). As a consequence, no statistically significant temporal variations of antimicrobial susceptibility rates occurred during the four-year study time

A four-year trend of in vitro sensitivity profile of Enterococcal strains (years 2004-2007)

The temporal trend of the in vitro antibiotic susceptibility rates was examined for all Enterococcus faecalis and Enterococcus faecium strains, isolated at our General Teaching Hospital during the years 2004-2007. Among Enterococcus faecalis isolates (2,736 strains tested on the whole), the greater activity rate was achieved by linezolid (100% of tested strains), followed by teicoplanin (97.9% to 100% of tested strains), nitrofurantoin (96.4% to 98.3% of isolates), vancomycin (81.0% to 100%), ampicillin (90.2% to 91.9%), penicillin (88.8% to 91.5%), while irregular variations of sensitivity occurred over time for gentamicin (however above 60% of tested strains), streptomycin (above 70% of overall strains), and tetracyclines (less than 20% of tested strains). When considering Enterococcus faecium strains (626 overall isolates), only linezolid maintained a 100% in vitro activity, followed by teicoplanin (87.7% to 100% of tested strains), vancomycin (78.4% to 86.2% of isolates), tetracyclines (56.8% to 81%), and gentamicin (59.1% to 71.0%), while an unpredictable in vitro efficacy was shown by streptomycin (27.6% to 69.8% of tested strains). A globally increased in vitro resistance rate was also detected for tetracyclines, during the entire four-year study period (p<.01). Sixty-six overall strains of vancomycin-resistant enterococcal (VRE) strains were detected, with a clearly increased trend from the year 2004 (7 cases only), up to the year 2007 (21 cases) (p<.001).

Enterococci cultured in the year 2007: an update

In vitro antibiotic susceptibility rates were prospectively investigated for all Enterococcus faecalis and Enterococcus faecium strains, isolated at our tertiary-care Hospital during the year 2007, which have been reported every three month in order to monitorize their epidemiological, clinical, and therapeutic concerns. Among Enterococcus faecalis isolates (705 strains tested on the whole), the greater antibiotic activity rate was achieved by linezolid (100% of tested strains), followed by nitrofurantoin (97.4% to 100% of isolates), teicoplanin (94.8% to 100%), vancomycin (87.5% to 100%), ampicillin (89.0% to 92.4%), penicillin (87.9% to 91.0%), while appreciable, but irregular variations of sensitivity occurred over time for gentamicin, streptomycin, and tetracyclines. With regard to Enterococcus faecium strains (175 strains), both linezolid and teicoplanin maintained a 100% in vitro activity, followed by vancomycin (86.4% to 100% of strains), streptomycin (62.5% to 100% of isolates), gentamicin (52.9% to 63.6%), and tetracyclines (51.8% to 49.6%), while negligible efficacy was shown by ampicillin (7.5% to 18.5% of tested strains), and penicillin (7.5% to 18.5%). Eighteen strains of vancomycin-resistant Enterococcus faecalis strains were detected (12 concentrated in the July-September period), while vancomycin-resistant Enterococcus faecium strains were six through the entire observation year (2007). No significant temporal modifications of antimicrobial sensitivity rates were observed, as well as no significant change in the emergence of vancomycin-resistant strains during time.

Escherichia coli and other Enterobacteriaceae (years 2004-2007)

The continued modification of drug resistance rates among Enterobacteriaceae is a relevant issue, especially in hospital facilities. The temporal variations of the in vitro antimicrobial sensitivity trends were updated quarterly for all suitable Enterobacteriaceae strains, followed from January 1, 2004, up to December 31, 2007. Among overall Escherichia coli isolates (4,413 strains tested on the whole), the carbapenem imipenem and colistin maintained a full (100%) in vitro activity, followed by amikacin (97.3% to 99.5% of tested strains), nitrofurantoin (89.2% to 94.8%), piperacillin-tazobactam (89.0% to 93-9%), gentamicin (81.9% to 89.4%), ceftazidime (78.3% to 89.5%), cefotaxime (78.0% to 89.8%), and ciprofloxacin (63.8% to 73.9%). When considering Enterobacteriaceae other than Escherichia coli, imipenem and colistin remained 100% active, followed by amikacin (94.9% to 97-2% of tested strains), piperacillin-tazobactam (78.4% to 86.0%), cotrimoxazole (72.1% to 78.0%), gentamicin (74.0% to 77.3%), norfloxacin (66.4% to 76.2%), ceftazidime (62.7% to 69-7%), and cefotaxime (62.3% to 69.0%). The emerging spread of enlarged-spectrum beta-lactamase (ESBL) production among this microbial pathogens significantly reduced the activity of third-generation cephalosporins over time (from a mean of 89.7% of susceptible Escherichia coli strains in the year 2004, to 78.1% in the year 2007; p<.001; and from a mean of 69.7% of sensitive Enterobacteriaceae strains in the year 2004, to 62.4% in the year 2007; p<.03). Also fluoroquinolones and beta-lactam molecules protected by suicide bet-lactam inhibitors suffered from a drop of their in vitro sensitivity rates (p<.02 to p<.005).

Pseudomonas aeruginosa (2004-2007)

Particular attention has been deserved to Pseudomonas aeruginosa, as a leading Gram-negative organism, which often tests multiresistant especially among inpatients and in hospital settings. The temporal variations of the in vitro antimicrobial sensitivity rates of all isolated Pseudomonas aeruginosa strains were collected for all suitable isolates,during the four-year period ranging from January 2004, up to December 2007. Among overall Pseudomonas aeruginosa isolates (2,083 evaluable tested strains), the best performance was obtained by the old colistin (colimycin), with a sustained 100% in vitro susceptibility rate, followed by amikacin (72.8% to 81.2% of tested strains), imipenem (76.8% to 80.8%), piperacillin- tazobactam (70.9% to 78.7%), ceftazidime (68.9% to 77.1%), and tobramicin (64.6% to 70.9%). On the other hand, gentamicin (55.1% to 63.5% of tested strains), aztreonam (57.5% to 66.8%), ciprofloxacin (55.7% to 65.0%), ticarcillin- clavulanate (53.7% to 60.5%), and mezlocillin (48.4% to 55.2%), proved less affordable in their in vitro antibacterial activity, as potential compounds to be administered on empiric basis (i.e. pending in vitro suseptibility assays). A significant temporal trend towards a reduced antibiotic sensitivity was found for the majority of tested molecules, but it resulted significant only for aztreonam and ciprofloxacin (p<.001 each), ticarcillin-clavulanate (p<.02), as well as for mezlocillin and tobramicin (p<.04 each).

Pseudomonas aeruginosa cultured in the year 2007: an update

The temporal variations of the in vitro antimicrobial sensitivity rates were registered at quarterly intervals for all suitable P. aeruginosa strains, detected at our Hospital during the last year 2007. Among P. aeruginosa isolates (493 strains tested on the whole), the best performance was obtained by the old colistin (colimycin), with a 100% susceptibility rate, followed by piperacillin-tazobactam (79.3% to 81.4% of tested strains), amikacin (72.8% to 82.6% of isolates), imipenem (73.0% to 85.2% of tested strains), ceftazidime (from 75.6% to 82.9% of tested strains), tobramycin (from 57.0% to 77.9% of microbial isolates). On the other hand, significantly less effective sensitivity profiles were shown by gentamicin (52.2% to 71.8% of tested strains), ciprofloxacin (51.0% to 65.7% of tested strains), aztreonam (50.8% to 62.2% of isolates), ticarcillin- clavulanate (49.1% to 59.5%), and mezlocillin (49.1% to 54.2% of tested strains). When examining temporal trends of antibiotic sensitivity figures, significant changes were observed only for ceftazidime and ciprofloxacin (p<.03) (with their drop of susceptibility, essentially due to the frequent production of ESBL by microbial strains).

Extended-spectrum beta-lactamase production (year 2007)

The increased rate of antimicrobial resistance among Gramnegative rods and all Enterobacteriaceae as a whole is a major concern, especially in the hospital setting. The prospective microbiological surveillance of antimicrobial susceptibility rates ongoing at our Hospital, has been implemented during the year 2007 with an analysis of the relationship with the emergence of extended-spectrum beta-lactamase (ESBL) secretion. Among Escherichia coli isolates (493 strains tested on the whole), imipenem proved in vitro effective in 100% of cases, followed by piperacillin-tazobactam (86.6% to 90.6% of tested strains), nitrofurantoin (86.1% to 91-7% of strains), ceftazidime (75.8% to 79.8%), cefotaxime (75.7% to 79.8%), co-amoxiclav (61.7% to 69.8%), ciprofloxacin (61.0% to 64.3%), and norfloxacin (64.5% to 70.2%). Both cefotaxime and ceftazidime sensitivity (seriously affected by the production of ESBL), showed a drop from a 79.8% mean susceptibility rate observed during the first quarter of year 2007, to a mean 75.7% of the last quarter of the same year 2007. With regard to Enterobacteriaceae as a whole (other than Escherichia coli), among 753 comprehensive isolates, both imipenem and colistin retained full (100%) in vitro activity, followed by piperacillin-tazobactam (72.5% to 81.5% of tested strains), cotrimoxazole (71.6% to 77.6%), gentamicin (71.6% to 76.1%), ciprofloxacin (64.1% to 68.6%), ceftazidime (60.8% to 64.5%), norfloxacin (60.2% to 69.8%), and cefotaxime (59.7% to 62.7%), with cephalosporins moderately affected by ESBL production, although in absence of significant temporal modifications.

Discussion

An active, prospective bacteriological-mycological surveillance project may notably add to the knowledge of local epidemiological figures and in vitro antimicrobial sensitivity trends, therefore playing a major role when planning surveillance measures, chemoprophylaxis regimens, and empiric antimicrobial treatment, especially on local and regional basis2-4,7,9-12, among both adult and pediatric patients13. A prospective microbiological monitoring with periodic reports (quarterly web-accessible and hospital-shared reports in our case), is expected to significantly contribute in the selection of both chemoprophylaxis and treatment choices of antimicrobial compounds, contributing to contain the spread of microbial resistance among inpatients, indirectly in the hospital environment, and secondarily in the community.

Although the major causative agents of inpatient infections remain the Enterobacteriaceae, however a significant decline of coagulase-negative Staphylococci, all Streptococci, and Gram-negative oxydase-negative organisms occurred over the examined four-year period, while Enterococci showed a mild increase over time in our experience.

Special attention should be deserved to bacteremias and sepsis, which are responsible for considerable morbidity and mortality rates among inpatients6,7,11,13. Although the main ethiological agents of inpatient bacteremias are still represented by coagulase-negative Staphylococci, these microorganisms significantly declined in thir overall figures during our four-year study period, thus confirming a positive trend towards a progressively reduced incidence of contaminated blood cultures. On the other hand, an appreciable increased frequency occurred over time for Pseudomonas, Klebsiella, and Enterococcus spp., as already noticed6,13. A major, persisting role as an agent of hospital bacteremic episodes was still exerted by E. coli among Gram-negative pathogens, and by S. aureus among Gram-positive ones.

Central vascular catheter (CVC) infections represent relevant causes of hospital-related morbidity and mortality6,14,15, and their careful monitoring plays a highly significant role in the selection and planning of chemoprophylactic and therapeutic choices in high-risk settings like surgical and intensive care units, on both local and regional settings. Although the major causative agents of CVC-related infection among hospitalized patients remain Staphylococci as a group (as already known)6,14,15, however the progressive emerging of Gramnegative pathogens is appreciable also over a proportionally short (12-month) observation period, and deserves major attention by Microbiologists and Clinicians involved in clinical care with specific competences.

On the other hand, urinary tract infections are responsible for considerable morbidity among a large number of inpatients, burdened by multiple risk factors for these infections10,16-18, some of them deserving accurate monitoring and selection of appropriate devices16-18. During the still limited observation time of our study, modest variations occurred in the frequency of isolation of the most frequent microorganisms, with Escherichia coli representing nearly 40% of cultured organisms, followed by E. faecalis (which proved responsible of around 23% of overall episodes). The tendency towards an increased incidence of E. faecium as an agents of urinary tract diseasee is of great concern, given the unpredictable antibiotic sensitivity profile of this last emerging Gram-positive microorganism18.

When considering the microbiological perspective, seen through the main microbial pathogens, S. aureus seems to play a steady, significant role among inpatients, due to its frequency, pathogenicity, broad range of clinical features, and elevated frequency of antimicrobial resistance rates19. Methicillin resistance is a well known marker of a wide range of antimicrobial resistance in these organisms, so that its is carefully monitorized through contries and with supranational projects6,7,9,19. In our experience, despite a stable, significant rate of methicillin resistance rate (mean value around 47% of all S. aureus isolates in a four-year period), we have to underline that “older” antimicrobial compounds like cotrimoxazole, chloramphenicol, and also rifampicin, may still play some role in selected clinical situations20 (possibly as add-on agents), while the activity of both available glycopeptides was completely preserved in our experience.

With regard to Enterococci as a group, the emerging of in vitro resistance to some reference compounds, and the initial record of some sparse vancomycin-resistant organisms (VRE) in particular, may be adequately monitorized on temporal basis as planned by our study, in order to address the clinical choices according to the local epidemiology and antimicrobial testing features. The emerging of in vitro resistance to some reference compounds like glycopeptides may be also well targeted on these basis, in order to preserve the clinical use of the majority of molecules which still guarantee effective activity of these difficult-to-treat Gram-positive cocci (i.e. novel agents with extended activity against multiresistant Gram-positive cocci, like linezolid, quinupristin-dalfopristin, dalbavancin, oritavancin, telavancin, ceftobiprole, and ceftaroline, for instance)21.

When considering E. coli and the vast spectrum of Enterobacteriaceae, despite a maintained activity of all carbapanems and that of the old colistin, a significant trend towards increased antimicrobial resistance rates was found over a four-year observation period, with extended-spectrum betalactamase (ESBL) secretion playing a major role in their resistance pathways, as already shown1,12,17,22-24.

A prospective monitoring of antimicrobial susceptibility rates of a major hospital-associated and environmental organism like P. aeruginosa is relevant, to add to local and national guidelines of antibiotic treatment and prophylaxis7,25. Despite a progressive, significant increase of resistance rates against the majority of compounds which usually test active against P. aeruginosa, however amikacin, carbapanems, piperacillintazobactam, amikacin, and ceftazidime still maintain a reliable role in eventual, empiric regimens to be added pending microbial isolation and in vitro sensitivity assays, since they remained active in at least 70% of hospital isolates of the last four examined years (2004-2007). The old polymixine colistin, which maintains full in vitro activity against all Pseudomonas strains, remains as a possible component of combined antimicrobial strategies1,7,20, or synergistically acting compounds7,25, when multiresistant pathogens are of concern. During the last year of observation (2007), despite a significant increase of resistance rates of P. aeruginosa against ceftazidime and ciprofloxacin (attributable to a rise of spread of ESBL secretion by Pseudomonas spp. too), however piperacillin-tazobactam, amikacin, imipenem, and the same ceftazidime still maintain a reliable role in eventual, empiric regimens to be added pending microbial isolation and in vitro susceptibility studies, since they remained active in nearly 80% of hospital isolates of the last study year (2007).

Finally, the increased rate of in vitro antimicrobial resistance among Gram-negative rods and all Enterobacteriaceae as a whole represents a major concern, especially in the hospital setting and among compromised inpatients. An analysis of the relationship with the emergence of ESBL secretion also confirmed the major role of this last mechanism in the spread of resistance determinants1,22-24,26. Prospective surveillance studies of the in vitro antimicrobial sensitivity rates of some relevant hospital-associated organisms like Escherichia coli and Enterobacteriaceae are an useful guidance to plan costeffective antibiotic treatment and prophylaxis, on local and regional basis, also taking into account the environmental perspectives of antibiotic usage4,6,7,22-26. These last Gram-negative organism groups also allow a reliable study of the temporal trend of ESBL production, since this feature significantly affects the activity of multiple broad-spectrum antimicrobial compounds against these commonly isolated Gram-negative microorganisms. Among them, and increasing concern is focused on organisms like Acinetobacter spp., which test resistant to almost all antimicrobials directed against Gramnegative rods27,28, and often require an association of the old colistin with carbapenems, tigecycline, rifampicin, or other agents carefully selected on the groud of extended in vitro sensitivity testing.

In conclusion, the reported figures and their continued update and discussion on the ground of the most recent literature evidence, underline the importance of well-constructed epidemiological studies to determine the prevalence of antimicrobial resistance in daily clinical practice, and the central place of reference laboratory-based in vitro sensitivity testing in predicting antimicrobial chemotherapy and subsequently in optimizing patient outcomes. As health care systems vary widely, planning and prevention strategies must be designed accordingly, although economics in general, cost factors, and resource planning are becoming common issues to almost all countries.

References

1. Nordmann P, Naas T, Fortineau N, Poirel L. Superbugs in the coming new decade: multidrug resistance and prospects for treatment of Staphylococcus aures, Enterococcus spp., and Pseudomonas aeruginosa in 2010. Curr Opin Microbiol 2007; 10:436-440.        [ Links ]

2. Macgowan AP. BSAC Working Parties on Resistance Surveillance. Clinical implications of antimicrobial resistance for therapy. J Antimicrob Chemother 2008; 62 (Suppl. 2):ii105-ii114.        [ Links ]

3. Marcel JP, Alfa M, Baquero F, Etienne J, Goossens J, Harbarth S, et al. Healthcare-associated infections: think globally, act locally. Clin Microbiol. Infect 2008; 14:895-907.        [ Links ]

4. Masterton R. The importance and future of antimicrobial surveillance studies. Clin Infect Dis 2008; 47 (Suppl. 1):S21-S31.        [ Links ]

5. Hawkey PM. The growing burden of antimicrobial resistance. J Antimicrob Chemother 2008; 62 (Suppl. 1):i1-i9.        [ Links ]

6. Casey AL, Mermel LA, Nightingale P, Elliott TS. Antimicrobial central venous catheters in adults: a systematic review and meta-analysis. Lancet Infect Dis 2008; 8:763-776.        [ Links ]

7. Maragakis LL., Perencevich EN, Cosgrove SE. Clinical and economic burden of antimicrobial resistance. Expert Rev Anti Infect Ther 2008; 6:751-763.        [ Links ]

8. Clinical and Laboratory Standards for Antimicrobial Susceptibility Testing (AST), 2008. Clinical Laboratory Standards Isntitute (CLSI). Update available on web: www.clsi.org        [ Links ]

9. Stefani S, Varaldo PE. Epidemiology of methicillin-resistant staphylococci in Europe. Clin Microbiol Infect 2003; 9:1179-1186.        [ Links ]

10. Johansen TE, Cek M, Naber KG, Startchounski L, Svendsen MV, Tenke P. Hospital acquired urinary tract infections in Urology departments: pathogens, susceptibility and use of antibiotics. Data from the PEP and PEAP study. Int J Antimicrobial Agents 2007; 28 (Suppl. 1):S91-S107.        [ Links ]

11. Schonheyder HC, Sogaard M. Hospital-acquired bacteraemia and fungaemia. A regional study with national implications. Ugeskr Laeger 2007; 169:4175-4179.        [ Links ]

12. Esposito S, Fiore M. Use of fluoroquinolones in intensive care units in a multi-drug-resistant era. Infez. Med 2007; 15:159-168.        [ Links ]

13. Pérez-González LF, Ruiz-González JM, Noyola DE. Nosocomial bacteremia in children: a 15-year experience at a general Hospital in Mexico. Infect Control Hosp Epidemiol 2007; 28:418-422.        [ Links ]

14. Viale P, Stefani S. Vascular catheter-associated infections: a microbiological and therapeutic update. J Chemother 2006; 18:235-249.        [ Links ]

15. Zingg W, Cartier-Fässler V, Walder B. Central venous catheter-associated infections. Best Pract Res Clin Anesthesiol 2008; 22:407-421.        [ Links ]

16. Esposito S, Noviello S, Leone S. Catheter-associated urinary tract infections: epidemiology and prevention. Infez Med 2008; 16:130-143.        [ Links ]

17. Woodford HJ, George J. Diagnosis and management of urinary tract infection in hospitalized older people. J Am Geriatr Soc 2008 (November 14) [Epub ahead of print].        [ Links ]

18. Lichtenberger P, Hooton TM. Complicated urinary tract infections. Curr Infect Dis Rep 2008; 10:499-504.        [ Links ]

19. Moreillon P. New and emerging treatment of Staphylococcus aureus infections in the hospital setting. Clin Microbiol Infect 2008; 14 (Suppl. 3):32-41.        [ Links ]

20. Falagas ME, Kopterides P. Old antibiotics for infections in critically ill patients. Curr Opin Crit Care 2007; 13:592-597.        [ Links ]

21. Arias CA, Murray BE. Emergence and management of drug-resistant enterococcal infections. Expert Rev Anti Infect Ther 2008; 6:637-655.        [ Links ]

22. Cornaglia G, Akova M, Amicosante G, Cantón R, Cauda R, Docquier JD, et al. Metallo-beta-lactamases as emerging resistance determinants in Gram-negative pathogens: open issues. Int J Antimicrob Agents 2007; 29:380-388.        [ Links ]

23. Pitout JD, Laupland KB. Extended-spectrum beta-lactamase-producing Enterobacteriaceae: an emerging public health concern. Lancet Infect Dis 2008; 8:159-166.        [ Links ]

24. Bassetti M, Repetto E. Diagnostic and therapeutic management of Gram-negative infections. Infez Med 2008; 16 (Suppl. 2):22-29.        [ Links ]

25. Bassetti M, Righi E, Viscoli C. Pseudomonas aeruginosa serious infections: mono or combination antimicrobial therapy?. Curr Med Chem 2008; 15:517-520.        [ Links ]

26. Rodríguez-Baòo J, Pascual A. Clinical significance of extended-spectrum beta-lactamases. Expert Rev Anti Infect Ther 2008; 6:671-683.        [ Links ]

27. Gilad J, Carmeli Y. Treatment options for multidrug-resistant Acinetobacter species. Drugs 2008; 68:165-189.        [ Links ]

28. Bassetti M, Repetto E, Righi E, Boni S, Diverio M, Molinari MP, et al. Colistin and rifampicin in the treatment of multidrug-resistant Acinetobacter baumanii infections. J Antimicrob Chemother 2008; 61:417-420.        [ Links ]