REPORT. INVESTIGATOR. R. Vanhoof 1, WIV/ISP Pasteurinstituut - Brussel, Eenheid Antibiotica-Onderzoek, Engelandstraat 642, B-1180 Brussel.

Transcription

1 REPORT. In vitro study to investigate the antimicrobial activity of various antibiotics against noninvasive clinical isolates of Streptococcus pneumoniae collected in Belgium during winter 7-8 (SP8). INVESTIGATOR. R. Vanhoof, WIV/ISP Pasteurinstituut - Brussel, Eenheid Antibiotica-Onderzoek, Engelandstraat 6, B-8 Brussel. «No part of this report - including graphs, figures, tables may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing form from an authorized person representing the Scientific Institute of Public Health» The nd of September 8

2 INTRODUCTION. S. pneumoniae, with its high degree of morbidity and its considerable rate of mortality, is one of the most commonly identified pathogens both in community and hospital acquired infections. In 967, the first penicillin-insusceptible S. pneumoniae strain was reported from Australia and since then antibiotic resistance in this micro-organism became a problem of paramount importance. High rates of insusceptibility to penicillin and other related and unrelated compounds have been reported world-wide. Over the years, penicillin insusceptible isolates became more and more concomitantly resistant to other classes of antimicrobials. This can be attributed to the de novo acquisition of new genetic material and the clonal spread of resistant isolates. In Belgium, the first two penicillin resistant isolates were reported by Vanhoof et al in 985. The situation had been evolving less spectacular but since the beginning of the nineties both the Antibiotic research Unit of the Institute Pasteur of Brussels and the National Reference Centre have been reporting a slow but steadily increase in penicillin insusceptibility in clinical isolates. In the former surveillance studies, we reported a level of penicillin insusceptibility of.5,.3, 6.,., 5.,.7, 5.,.8 and. for winter 995, 997, 999,, 3,, 5, 6 and 7 respectively. The epidemiology of antibiotic resistance can be influenced by various parameters and geographic differences, even on a regional scale, have been reported. These findings underline the necessity for continuous national and international surveillance of resistance. Therefore we organized this th collaborative surveillance study, to monitor possible changes in antibiotic resistance in non-invasive clinical isolates of S. pneumoniae collected by 5 participating centres. MATERIAL AND METHODS. This study was conducted in collaboration with 5 selected clinical centres in Belgium. The aim of this eight annual surveillance study was to obtain information on the level and evolution of antibiotic resistance in Belgian isolates of S. pneumoniae. The isolates of Streptococcus pneumoniae prospectively collected were obtained from the following centres: Hôpital de la Citadelle, Liège. (Dr. M. Carpentier), Laboratoire Cebiodi, Bruxelles (Dr. B. Mulongo), Clinique Universitaire de Mont-Godinne, Yvoir (Dr. Y. Glupczynski), Clinique St. Joseph, Arlon (Dr. P. Goffinet), C.H.U. André Vésale, Montignies-le-Tilleul (Dr. D. Govaerts), Hôpital Princesse Paola, Marche-en-Famenne (Dr. Ph. Lefèvre), Medisch

3 Centrum Huisartsen, Leuven (Apoth. M. Lontie), Virga-Jesseziekenhuis, Hasselt (Dr. R.Cartuyvels), Hôpital de Jolimont, Haine-St.-Paul (Dr. F. Meunier), CHR Hôpital de Warquignies, Boussu (Dr. I. Philippart), H. Hartziekenhuis, Roeselare (Dr. I. Surmont), A.Z. Stuivenberg, Antwerpen ( Dr. K. Camps), A.Z. St. Jan, Brugge (Dr. B. Gordts), A.Z. Jan Palfijn, Gent (Dr. L. Ide) and Imeldaziekenhuis, Bonheiden (Dr. J. Frans). The following antibiotics were tested in the study and were provided as laboratory preparations with known potency: clavulanic acid, ceftazidime (GlaxoSmithKline), cefepime (Bristol Myers Squibb), cefotaxime, levofloxacin, ofloxacin and telithromycin (Sanofi Aventis), ciprofloxacin and moxifloxacin (Bayer), azithromycin (Pfizer). Amoxicillin, ampicillin, cefaclor, cefuroxime, clindamycin, erythromycin, penicillin G and tetracycline were obtained from a commercial source (Sigma). Imipenem was obtained from Fisher Scientific, USA. Amoxicillin/clavulanic acid was tested in a : ratio. All antibiotics were tested for 6 serial twofold dilutions (. 3 µg/ml). The Minimal Inhibitory Concentrations (MICs) were determined by broth microdilution as recommended by the CLSI. All isolates underwent a slide agglutination (Slidex pneumo Kit TM, BioMérieux) and an Optochine test (OPTO-F, BioMérieux) before MIC testing. All the isolates were also tested for the LytA gene by PCR. S. pneumoniae ATCC 969, S. pneumoniae TPN88 (internal control isolate) and Staphylococcus aureus NCTC 56 (ß-lactamase positive to validate the clavulanate component of amoxicillin/clavulanate) were included as quality control organisms in each series. Interpretation of the results was based on breakpoints provided by the CLSI. Levels of susceptibility to ampicillin were determined by the penicillin breakpoints. For ciprofloxacin, we used breakpoints of one dilution lower than those of levofloxacin as is generally the case for other types of micro-organisms. The penicillin insusceptible isolates were typed by the National Reference Centre by using the Quellung reaction with sera from the Staten Serum institute (Copenhagen, Denmark). The Chi-square test, with or without Yates correction, for two independent samples was used for the statistical evaluation of the results. The level of significance was set at.5. RESULTS. In total 8 documented isolates of S. pneumoniae were included in the study for further analysis. Seventeen point four percent (78/8) of the isolates were from children (age 5 years) with 66/78 or 8.6 from children under 5 years of age, while 8.6 (37/8) were from adults with 38/37 or 6.3 from adults with age 6 years. Age showed a 3

4 bimodal distribution with a first peak between and 5 years (.7 ) and a second peak between 6 and 8 year (5. ). The mean age of the study population was 53. years. The mean age per collection centre varied from 38. years (Medisch Centrum Huisartsen, Leuven) to 73.7 (Hôpital Princesse Paola, Marche-en-Famenne). Isolates from sputum represented 79.9 (358/8) of the specimens, 5. (69/8) were from nasal swab,.5 (/8) from throat,.8 (8/8) from sinus and. (/8) from respiratory pus. Overall, 79.9 the isolates were from lower respiratory tract (LRT) specimens and. from upper respiratory tract (URT). Isolates from sputum were significantly more present in patients of the age group 6 years (96. 9/38) when compared to the other age groups: -5 years (. ; 8/66; P<.), 6-5 years (58.3 ; 7/; P <.) and 6-59 years (86. ; /3;.> P >.). Isolates from sputum were also more present in the age group 6-59 years (86. ; /3) than in the age groups -5 years (. ; 8/66; P<.) and 6-5 years (58.3 ; 7/;.5 > P >.). Overall, this means that Sputum samples were significantly more present in adults than in children (95.8 versus.; P <.). Isolates from upper respiratory tract origin were significantly more present in children than in adults (7. versus 3. ; P <.). This was especially due to the age group -5 years in which 87.9 of the samples were from URT. Isolates from hospitalised patients represented 6.7 (8/8) of the isolates while 35.3 (58/8) were from ambulatory patients and. (9/8) from long care facility patients. Sixty four point one percent (87/8) of the isolates were from male patients. Isolates from the Northern part of the country represented 6.7 (9/8) while 5.5 (/8) and 7.8 (35/8) of the isolates came from patients from the Southern part and Brussels respectively. The highest intrinsic activity on a weight basis was found for imipenem (MIC 5, or MIC for 5 of the isolates tested, of. µg/ml) followed by amoxicillin, amoxicillin/clavulanic acid, and telithromycin (MIC 5.8 µg/ml), penicillin G, ampicillin cefotaxim, ceftazidime, cefepime, (MIC 5.5 µg/ml), cefuroxime (MIC 5,3 µg/ml), moxifloxacin, erythromycin, azithromycin (MIC 5.6 µg/ml), tetracycline (MIC 5. µg/ml), cefaclor, clindamycin, (MIC 5.5 µg/ml), levofloxacin (MIC 5.5 µg/ml), ciprofloxacin and ofloxacin (MIC 5. µg/ml) (Table ). Telithromycin, amoxicillin, amoxicillin/clavulanic acid, moxifloxacin, levofloxacin, cefotaxime, ceftazidime and cefepime were the compounds with the highest level of susceptibility (,,, 99.6, 99.3, 98, 98, and 98 resp.) followed by imipenem, ciproflocacin, ofloxacin, cefuroximeaxetil and cefuroxime (97.5, 9.9, 9.9, 9.3 and 9.8 resp.).

5 5

6 The lowest degrees of susceptibilitiy were found for erythromycin, azithromcycin, clindamycin and tetracycline (7.3, 7.3, 73.7 and 78. respectively.) (Table ). TABLE : SUSCEPTIBILITY RATES FOLLOWING CLSI CRITERIA OF 8 NON- INVASIVE ISOLATES OF S. pneumoniae. (SP8) Antibiotic Susceptibility Rates (in ) Susceptible Intermediate Resistant Penicillin Ampicillin Amoxicillin Amoxicillin/clavulanate Cefaclor Cefuroxime Cefuroxime-axetil Cefotaxime 98.. Ceftazidime 98.. Cefepime 98.. Imipenem Ciprofloxacin Levofloxacin Moxifloxacin Ofloxacin Erythromycin Azithromycin Telithromycin Clindamycin Tetracycline See comment Text Cefuroxime-axetil: oral form of Cefuroxime Decreased susceptibility (I+R) to penicillin was.6 [7.6 intermediate (I =. - µg/ml) and. high-level (R µg/ml)], ampicillin. [7. I (=. - µg/ml) and. R ( µg/ml)], amoxicillin ± clavulanic acid [I = µg/ml and R 8 µg/ml], cefaclor.3 [.7 I (= µg/ml) and 9.6 R ( µg/ml)], cefuroxime 9. [.5 I (= µg/ml) and R ( µg/ml)], cefuroxime-axetil 8.7 [.9 I (= µg/ml) and 7.8 R ( µg/ml)], cefotaxime, ceftazidime and cefepime. [. I (= µg/ml) and (R µg/ml)], imipenem.5 [.5 I (=.5.5 µg/ml) and R ( µg/ml)], ciprofloxacin 5. [.7 I (= µg/ml) and. R ( µg/ml)], levofloxacin.7 [. I (= µg/ml) and.5 R ( 8 µg/ml)], moxifloxacin. [. I = µg/ml and. R = µg/ml)], ofloxacin 5. [.7 I (= µg/ml) and. R ( 8 µg/ml)], erythromycin 9.7 [.5 I (=.5 µg/ml) and 9. R ( µg/ml)], azithromycin 9.7 [.9 I (= µg/ml) and 8.8 R ( µg/ml)], telithromycin [I = µg/ml and R 6

7 µg/ml)], clindamycin 6.3 [.9 I (=.5 µg/ml) and 5. R ( µg/ml)] and tetracycline.9 [5. I (= µg/ml) and 6.5 R ( 8 µg/ml)] (Table ). In general, isolates showing resistance to an antibiotic (IR-isolates) were more present in children (36/78, 6. ) than in adults (3/37; 35.7 ) though the difference was not significant. Furthermore, there was no significant difference between the different age groups as far as the presence of IR-isolates is concerned. Resistance to erythromycin was significantly higher in children (.3 ; 33/78) than in adults (7.3 ; /37) (.>P>.). Resistance in the age group -5 year (3.9 ; 9/66) was also significantly higher than in the age groups 6-59 years (.; 3/3,.>P>.) and 6 year (9.; 69/38;.5>P>.). For Tetracycline there was a significant difference between the age group -5 year (3.8; /66) and 6-59 year (8.; /3) (.5>P>.). No further significant differences were found in the various age groups. Table 3 : of isolates in different age groups with resistance to age Pen Cip Ery Tet IR S Children Adults total Fig. : Resistance Rates: Children versus Adults (8) P C E*** T R (*Significant difference) Child Adult 7

8 Table : Resistance rates () in URT and LRT isolates (8) Pen Cip Ery Tet IR S URT LRT TOT Fig. : Resistance rates in URT and LRT (8) URT LRT 5 P C E T IR Isolates showing antibiotic resistance were more prevalent in URT (. ) than in LRT (36.9 ) though the difference was statistically not significant. No significant differences were found for penicillin, ciprofloxacin, erythromycin and tetracycline. Fig.3: Resistance rates by region (8) P* C E T IR North South Bruss In general, the presence of isolates with decreased antibiotic susceptibility revealed to be comparable in the three different regions: Brussels (. ), the Southern part (38.7 ) and the Northern part (35.9 ). The only significant difference was found for penicillin between Brussels (.9, 8/35) and the Northern part (8. ; 7/9) (.>P>.). No other significant differences were found for the different antimicrobials. 8

9 Table 5 : Resistance rates () following gender and type of isolate Pen Cip Ery Tet IR S Gender Male Female Type AMB HOSP LCF AMB:Ambulatory; HOS:Hospitalized; LCF:Long Care Facility Tetracycline resistance was significantly higher in males (5. ; 73/87) than in females (5.5 ; 6/6) (Table 5). As far as the admission type (ambulatory, hospitalized) concerned, no significant differences were found (Table 5). The most common resistance phenotypes (Table 6) were insusceptibility to Erythromycin-Tetracycline (.7 ), isolated insusceptibility to Erythromycin (7. ) followed by insusceptibility to Penicillin-Erythromycin-Tetracycline (5.8 ). TABLE 6 : DISTRIBUTION OF THE PENICILLIN-CIPROFLOXACIN- ERYTHROMYCIN-TETRACYCLINE SUSCEPTIBILITY PHENOTYPES (8). Penicillin-Erythromycin-Tetracycline-Ciprofloxacin phenotype Number () Susceptible 8 (6.5) Ery-Tet 57 (.7) Ery 33 (7.) Pen-Ery-Tet 6 (5.8) Cip 5 (3.) Pen (.) Pen-Ery (.) Tet 9 (.) Pen-Cip-Ery-Tet (.9) Cip-Ery-Tet (.5) Pen-Cip (.) Pen-Cip-Ery (.) Insusceptibility to one compound was present in 5. of the isolates. Two-, three- and fourfold resistance was found in 5., 6.5 and.9 respectively of the isolates. 9

10 Fig.: Distribution of Resistance Phenotypes 6,5,9 5, 5, 6,5 S AB AB 3AB AB MICs of all ß-lactams rose with those of penicillin. Overall, penicillin, ampicillin, Cefotaxime, ceftazidime and cefepime were equally active against the penicillin insusceptible isolates. Imipenem was generally three doubling dilutions more potent on a weight basis, while amoxicillin, amoxicillin/clavulanate were generally one doubling dilution more potent. Cefuroxime and Cefuroxime-axetil were two dilutions less active while cefaclor was 6 dilutions less active. Cross-resistance between penicillin and the other ß-lactams was not complete. The results indicate that of the penicillin-insusceptible isolates remained susceptible to amoxicillin ± clavulanate while 8.7 of the penicillin-insusceptible isolates remained susceptible to cefotaxime, ceftazidime, cefepime and 78.8 to imipenem (Table 7).

11 TABLE 7 : MIC VALUES OF ß-LACTAMS FOR VARIOUS PENICILLIN SUSCEPTIBILITY CATEGORIES (8). ß-Lactam PEN S (n = 396) PEN I/R (n = 5) MIC5 MIC9 MIC5 MIC9 S Penicillin Ampicillin Amoxicillin Amoxicillin/Clavulanate Cefaclor Cefuroxime Cefuroxime-axetil Cefotaxime Ceftazidime Cefepime Imipenem Oral form of cefuroxime The most important capsular types in penicillin-insusceptible isolates were capsular types 9 (5. ), (9.3 ), 3 (5.) 5 (3.5) 6 (9.6) and 9 (7.7 ).Other types were capsular type 7,, 9, 3 and 35 (each.9 ). Comparison with former surveys. Resistance rates. Table 8 summarizes the resistance rates for the different antimicrobial compound tested in the various surveillance studies conducted between 995 and 8. The graphic representations of the evolution for the most important classes of compounds are given in Fig. 5, 6 and 7. Penicillin G peaked in (.) and there is a clear downwards tendency in resistance to notice after with an actual resistance rate of.6. This resistance rate is in fact lower than the rate found in 995 (.5) at the start of the surveillance programme. The other beta-lactams followed this evolution though some minor shifts in time can be found. Ciprofloxacin peaked in 999 (5.) and decreased systematically since then. Levofloxacin had its peak of resistance in 3 (3.3) and followed the downward tendency of Ciprofloxacin. The resistance rate of Moxifloxacin remained low between and.6 (3). Rates for Macrolides (MLS) and Tetracycline ware fluctuating in time.

12 Table 8 : Non-susceptibility rates (IR) obtained in the various surveys PEN AMP AMX AMC CFC CRX CRXa CTX CTZ NT NT NT NT NT CPM NT NT NT NT NT IMI CIP LEV MOX OFL ERY AZI NT NT NT TEL NT NT NT NT.. CLI NT NT TET Fig. 5: Evolution of Resistance in Beta-Lactams PEN AMX CRX CTX IMI Fig 6: Evolution of Resistance in Fluoroquinolones CIP LEV MOX

13 Fig, 7: Evolution of Resistance in MLS and TET ERY TEL TET Interestingly, the absolute number of isolates showing high level resistance (MIC µg/ml) to Penicillin G decreased between and 7 (6.,.9,.8 and.7), though a slight increase in percentage was noted in 8 (.). In any case, the intermediate isolates (. µg/ml to. µg/ml) outnumbered the isolates with high level resistance to Penicillin G in the period -8 (Fig.8). In 997, 999 and 3, the number of high level isolates was more important than the intermediate isolates (6.7 vs 5.5; 9.7 vs 5. and 8. vs. 6.7 respectively). In 995 and, the numbers in both groups were nearly comparable. This indicates that there exists a general trend in lowering the number of isolates with high level resistance to Penicillin G. However, attention should be paid to the slight increase in 8. 3

14 Fig. 8: of isolates with I and R to Pen ,7 8, 6, 5,7 6,7 9,7, 6, 5,5 8, 6,7 5,,9,8,7,3 9, 7, R I As can been seen in Fig. 9, resistance to Ciprofloxacin and other fluoroquinolones is mainly due to the presence of isolates showing intermediate resistance. The only exception was 7. In this year, a slight predominance of high level isolates was seen though the difference between the two groups was very discrete. Fig. 9: of isolates with I and R to CIP 6 3,,8, 8 6 3,6,3,5 3,,3,7 7,7 6,6 9,5, 8, 6,6 6,,7, R I

15 Evolution of MIC distributions. Fig. : MIC distribution of PEN ,,,3,6,,5, Fig. shows clearly a bimodal distribution of the MIC for Penicillin G. This was also the case for the other beta-lactam antibiotics. This bimodal distribution was found in every survey. The bimodal character indicates that there exists a distinct modus for a susceptible population and another modus for the non-susceptible population. The Modus of the population was.5 µg/ml as it was the case since. This indicates that the population did not shift to higher MIC values. Fig. : Evolution of MIC5 in PEN-S and PEN-non S isolates PEN-S PEN-I PEN-R 5

16 Fig. shows the evolution of the MIC5 (expressed in µg/ml) values in Penicillin susceptible and Penicillin non-susceptible isolates collected during the various surveillance studies. The MIC5 can be considered as the indicator for intrinsic activity. The lower the value, the more active the compound is on a given population. A population becomes less sensitive to a certain compound when the MIC5 value shifts to the right side of the curve (i.e. the higher values). As it can be seen from this figure, the MIC5 increased importantly from 995 (.8 µg/ml) to 3 (.3 µg/ml), indicating that in this period the penicillin susceptible population became less susceptible. From, the MIC5 was positioned at a lower value i.e..5 µg/ml (with the exception of.8 µg/ml in 7). The MIC5 value for the non-susceptible did not vary significantly during the years. This means that the shift of the modus in the total population is due to changes in the susceptible population rather than to changes in the non-susceptible strains. Fig. : MIC distribution of CIP ,3,6,,5, Fig,3: MIC Distribution of LEV ,3,6,,5,

17 Fig,: MIC distribution of MOX ,,,3,6,,5,5 7 8 Fig,5: MIC distribution of OFL ,,5, Figures to 5 show the MIC distributions of Ciprofloxacin (CIP), Levofloxacin (LEV), Moxifloxacin (MOX) and Ofloxacin (OFL). As can been seen from these figures, the modus of the populations in the different years and for the different compounds did not fluctuate significantly. Furthermore, these compounds did not have a bimodal distribution. There distributions remained very Gaussian during the years. From figures 6, 7 and 8 it can be seen that for all the fluoroquinolones (Ofloxacine included but not shown as it is completely comparable to ciprofloxacin) the MIC5 value in both the susceptible and non-susceptible population did not shift in the various surveys. This indicates that the intrinsic activity of these compounds remained fairly stable during he period

18 Fig. 6: Evolution of MIC5 in CIP-S and CIP-non S isolates CIP-S CIP-I CIP-R Fig. 7: Evolution of MIC5 in LEV-S and LEV-non S isolates LEV-S LEV-I LEV-R 8

19 FIG. 8: Evolution of MIC5 in MOX-S and MOX-non S isolates MOX-S MOX-I MOX-R Comparison of Fluoroquinolone resistance rates following CLSI and EUCAST. Table 9: CLSI and EUCAST breakpoints for FQ CLSI EUCAST S I R S I R CIP,5,5- LEV 8 MOX,5 OFL 8,5,5-8 Table 9 compares the breakpoints of CLSI with those proposed by EUCAST. For EUCAST the wild type S.pneumoniae is not considered susceptible to Ciprofloxacin or Ofloxacin is therefore categorized as intermediate. The breakpoints for Levofloxacin relate to high dose therapy. Table and figures 9- show that the resistance rates for Levofloxacin and Moxafloxacin are very comparable if not identical with the two sets of breakpoints. On te other hand, there is an important difference for Ciprofloxacin and Ofloxacin (not shown inn figures), as the result of the categorization of the isolates as wild type for these two compounds. 9

20 Table : of non-susceptibiliy for Fluoroquinolones following CLSI and EUCAST breakpoints CLSI EUC CLSI EUC CLSI EUC CLSI EUC CLSI EUC CLSI EUC CIP 3,8 96, 9 96,7 7, 96, 8,6 98, 3,9 98,3 5, 99,3 LEV 3,3 3,3,8,8,6,6,5,5,9,7, MOX,5,5,,,,,,,7,5, OFL 3,5 99,5 9 98,8 7, 97,3 8, 99,5 3,6 99,5 5, 99,8 Fig. 9: CIP resistance rates following CLSI and EUCAST 96, 96,7 96, 98, 98,3 99,3 8 6 CLSI EUCAST 3,8 9 7, 8,6 3,9 5, Fig.: LEV resistance rates following CLSI and EUCAST 3,5 3,5 3,3,3,8,8,5,5,7,6,6,5,5,9, CLSI EUCAST Fig.: MO X resistance rates following CLSI and EUCAST,7,7,6,5,5,5,5,,3,,,,,,,, CLSI EUCAST,